Acta Universitatis Danubius. Œconomica, Vol 14, No 1 (2018)

Operations Research; Statistical Decision Theory

Analysis of Natural Movement of Romanian Population During 2007-2014 - IV

Cătălin Angelo Ioan¹

Abstract:. Article shall carry out the analysis of natural movement of Romanian population During 2007-2014. They are thus treated indicators: Live births, Deceased, Natural increase, Marriages, Divorces and Deaths under 1 year. In addition to the regression analysis, are determined the median, quartiles, the arithmetic mean and standard deviation for each indicator. Also the analysis examines dependence aforementioned indicators of regional GDP variation.

Keywords: Live births; Deceased; Natural increase; Marriages; Divorces

JEL Classification: Q56

1. Introduction

In what follows we shall carry out the analysis of natural movement of Romanian population During 2007-2014. They are thus treated indicators: Live births, Deceased, Natural increase, Marriages, Divorces and Deaths under 1 year. In addition to the regression analysis, are determined the median, quartiles, the arithmetic mean and standard deviation for each indicator. Also the analysis examines dependence aforementioned indicators of regional GDP variation.

In this fourth part, we shall analize the following counties: Satu Mare, Sibiu, Suceava, Teleorman, Timis, Tulcea, Valcea, Vaslui, Vrancea and entire country: Romania.

2. Analysis of Natural Movement of Romanian Population during 2007-2014

2.34. Analysis of Natural Movement of Satu Mare County Population

Statistics of natural movement corresponding to Satu Mare County are the following:

Table 199. The natural movement of Satu Mare County population during 2007-2008

Source: INSSE

Table 200. The natural movement of Satu Mare County population during 2009-2010

Source: INSSE

Table 201. The natural movement of Satu Mare County population during 2011-2012

Source: INSSE

Table 202. The natural movement of Satu Mare County population during 2013-2014

Source: INSSE

Table 203. The population trends of Satu Mare County during 2007-2014

Source: INSSE

Figure 364

From figure 364 we can see a sinusoidal evolution of the indicator. Except months aug 2007, sept 2007, iul 2008, aug 2008, sept 2008, sept 2009, aug 2010, aug 2011, aug 2012, sept 2012, iul 2013, aug 2013, iul 2014, aug 2014, sept 2014 the natural increase was negative.

Regression analysis relative to indicator “Live births” gives us an equation: y=-0.544641888x+332.9359649 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Deceased” gives us an equation: y=-0.423874118x+387.3078947 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Natural increase” gives us an equation: y=-0.12076777x+-54.37192982 where x is the number of month (Jan, 2007=1), therefore a downward trend.

For the set of values above, the median indicator for “Live births” is 303, for “Deceased” is 368 and for “Natural increase”: -63. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births”: (225,272.75,303,335,444), for “Deceased”: (275,338.75,367.5,388.25,466) and for “Natural increase”: (-175,-110,-62.5,-26.5,126).

The arithmetic mean and the standard deviation for “Live births” are: (307,46.16), for “Deceased”: (367,36.91) and for “Natural increase”: (-60,62.89). This means that with a probability greather than 0.68 “Live births” are in the range [261,353], for “Deceased” in [330,404] and for “Natural increase” in [-123,3].

Percentiles length indicators analysis (Figure 365) show that, indeed the concentration is around the middle of the data.

Figure 365

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Live births/10000 inh., Deceased/10000 inh. and Natural increase/10000 inh. as in the figure 366.

Figure 366

Regression analysis relative to indicator “Live births/10000 inh.” gives us an equation: y=-0.012813755x+8.376883772 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Deceased/10000 inh.” gives us an equation: y=-0.009635852x+9.747234649 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Natural increase/10000 inh.” gives us an equation: y=-0.003207474x+-1.368708333 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Live births/10000 inh.” is 8, for “Deceased/10000 inh.” is 9 and for “Natural increase/10000 inh.”: -2. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births/10000 inh.”: (5.69,6.895,7.7,8.4475,11.26), for “Deceased/10000 inh.”: (6.97,8.585,9.285,9.855,11.74) and for “Natural increase/10000 inh.”: (-4.43,-2.7725,-1.58,-0.67,3.21).

The arithmetic mean and the standard deviation for “Live births/10000 inh.” are: (8,1.16), for “Deceased/10000 inh.”: (9,0.92) and for “Natural increase/10000 inh.”: (-2,1.59). This means that with a probability greather than 0.68 “Live births/10000 inh.” are in the range [7,9], for “Deceased/10000 inh.” in [8,10] and for “Natural increase/10000 inh.” in [-4,0].

Percentiles length indicators analysis (Figure 367) show that, indeed the concentration is around the middle of the data.

Figure 367

A comparison of the indicator “Live births” with the national level shows that it is about the same with the national, being better in 47.92% cases. For “Deceased” the indicator is about the same with the national, being better in 54.17% cases. Finally, for “Natural increase”, the indicator is better than the national, being better in 61.46% cases.

Figure 368

Regression analysis relative to indicator “Marriages” gives us an equation: y=-0.929218665x+244.8796053 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Divorces” gives us an equation: y=-0.259407216x+59.8625 where x is the number of month (Jan, 2007=1), therefore a downward trend.

For the set of values above, the median indicator for “Marriages” is 161 and for “Divorces” is 46. Also, the distribution of quartiles is for “Marriages”: (54,114.75,161,244.75,609) and for “Divorces”: (8,33,46,60,109). The arithmetic mean and the standard deviation for “Marriages” are: (200,128.6) and for “Divorces”: (47,18.85). This means that with a probability greather than 0.68 “Marriages” are in the range [71,329] and for “Divorces” in [28,66].

Percentiles length indicators analysis (Figure 369) show that, indeed the concentration is around the middle of the data.

Figure 369

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Marriages/10000 inh. and Divorces/10000 inh. as in the figure 370.

Figure 370

Regression analysis relative to indicator “Marriages/10000 inh.” gives us an equation: y=-0.022860418x+6.163105263 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Divorces/10000 inh.” gives us an equation: y=-0.006421867x+1.507502193 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Marriages/10000 inh.” is 4 and for “Divorces/10000 inh.” is 1. Also, the distribution of quartiles is for “Marriages/10000 inh.”: (1.37,2.895,4.075,6.165,15.47) and for “Divorces/10000 inh.”: (0.2,0.83,1.17,1.5125,2.75). The arithmetic mean and the standard deviation for “Marriages/10000 inh.” are: (5,3.25) and for “Divorces/10000 inh.”: (1,0.48). This means that with a probability greather than 0.68 “Marriages/10000 inh.” are in the range [2,8] and for “Divorces/10000 inh.” in [1,1].

Percentiles length indicators analysis (Figure 371) show that, indeed the concentration is around the middle of the data.

Figure 371

A comparison of the indicator “Marriages” with the national level shows that it is about the same with the national, being better in 56.25% cases. For “Divorces” the indicator is about the same with the national, being better in 50% cases.

Figure 372

Regression analysis relative to indicator “Deaths under 1 year” gives us an equation: y=-0.041135377x+5.630482456 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year” is 3 and the distribution of quartiles is for “Deaths under 1 year”: (0,2,3,5,15). The arithmetic mean and the standard deviation for “Deaths under 1 year” are: (4,2.63) which means that with a probability greather than 0.68 “Deaths under 1 year” are in the range [1,7].

Percentiles length indicators analysis (Figure 373) show that, indeed the concentration is around the middle of the data.

Figure 373

Figure 374

Regression analysis relative to indicator “Deaths under 1 year/100000 inh.” gives us an equation: y=-0.010307311x+1.419175439 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year/100000 inh.” is 1 and the distribution of quartiles is for “Deaths under 1 year/100000 inh.”: (0,0.51,0.76,1.26,3.78). The arithmetic mean and the standard deviation for “Deaths under 1 year/100000 inh.” are: (1,0.66) which means that with a probability greather than 0.68 “Deaths under 1 year/100000 inh.” are in the range [0,2].

A comparison of the indicator “Deaths under 1 year” with the national level shows that it is about the same with the national, being better in 43.75% cases.

A final analysis examines dependence aforementioned indicators of regional GDP variation.

Table 204. The evolution of Satu Mare County GDP during 2007-2014

Source: INSSE and own calculations

In what follows, we shall investigate if there is a dependency between GDP variation (noted with dGDP) and the aforementioned indicators.

Searching dependence annual variations of “Live births” from GDP, we find that there is a dependence of Live births from GDP in the current year and the regression equation is: 0.8985dGDP+-2.0444. Searching dependence annual variations of “Deceased” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Natural increase” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Marriages” from GDP, we find that there is a dependence of Marriages from GDP offset by 1 year and the regression equation is:0.4977dGDP+-3.4798. Searching dependence annual variations of “Divorces” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Deaths under 1 year” from GDP, we find that there is not a dependence of the variation of GDP.

2.35. Analysis of Natural Movement of Sibiu County Population

Statistics of natural movement corresponding to Sibiu County are the following:

Table 205. The natural movement of Sibiu County population during 2007-2008

Source: INSSE

Table 206. The natural movement of Sibiu County population during 2009-2010

Source: INSSE

Table 207. The natural movement of Sibiu County population during 2011-2012

Source: INSSE

Table 208. The natural movement of Sibiu County population during 2013-2014

Source: INSSE

Table 209. The population trends of Sibiu County during 2007-2014

Source: INSSE

Figure 375

From figure 375 we can see a sinusoidal evolution of the indicator. Except months apr 2007, mai 2007, iun 2007, iul 2007, aug 2007, sept 2007, nov 2007, ian 2008, feb 2008, mar 2008, mai 2008, iun 2008, iul 2008, aug 2008, sept 2008, nov 2008, apr 2009, mai 2009, iun 2009, iul 2009, aug 2009, sept 2009, oct 2009, nov 2009, feb 2010, mar 2010, mai 2010, iun 2010, iul 2010, aug 2010, sept 2010, nov 2010, mar 2011, iun 2011, iul 2011, aug 2011, sept 2011, mai 2012, iun 2012, iul 2012, aug 2012, sept 2012, oct 2012, ian 2013, aug 2013, sept 2013, ian 2014, iun 2014, iul 2014, aug 2014, sept 2014, oct 2014 the natural increase was negative.

Regression analysis relative to indicator “Live births” gives us an equation: y=-0.259047748x+387.0846491 where x is the number of month (Jan, 2007=1), therefore a downward trend.

Regression analysis relative to indicator “Deceased” gives us an equation: y=0.055527672x+371.1506579 where x is the number of month (Jan, 2007=1), therefore an upward trend.

Regression analysis relative to indicator “Natural increase” gives us an equation: y=-0.314575421x+15.93399123 where x is the number of month (Jan, 2007=1), therefore a downward trend.

For the set of values above, the median indicator for “Live births” is 377, for “Deceased” is 368 and for “Natural increase”: 5. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births”: (282,343.5,377,395,507), for “Deceased”: (305,347.75,368,388.5,489) and for “Natural increase”: (-156,-34.25,5,42,152).

The arithmetic mean and the standard deviation for “Live births” are: (375,42.05), for “Deceased”: (374,37.88) and for “Natural increase”: (1,58.59). This means that with a probability greather than 0.68 “Live births” are in the range [333,417], for “Deceased” in [336,412] and for “Natural increase” in [-58,60].

Percentiles length indicators analysis (Figure 376) show that, indeed the concentration is around the middle of the data.

Figure 376

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Live births/10000 inh., Deceased/10000 inh. and Natural increase/10000 inh. as in the figure 377.

Figure 377

Regression analysis relative to indicator “Live births/10000 inh.” gives us an equation: y=-0.007125203x+8.483489035 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Deceased/10000 inh.” gives us an equation: y=-0.000327455x+8.137964912 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Natural increase/10000 inh.” gives us an equation: y=-0.00684475x+0.347907895 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Live births/10000 inh.” is 8, for “Deceased/10000 inh.” is 8 and for “Natural increase/10000 inh.”: 0. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births/10000 inh.”: (6.1,7.4925,8.24,8.595,11.05), for “Deceased/10000 inh.”: (6.59,7.5775,7.99,8.4275,10.61) and for “Natural increase/10000 inh.”: (-3.37,-0.745,0.11,0.915,3.31).

The arithmetic mean and the standard deviation for “Live births/10000 inh.” are: (8,0.92), for “Deceased/10000 inh.”: (8,0.82) and for “Natural increase/10000 inh.”: (0,1.27). This means that with a probability greather than 0.68 “Live births/10000 inh.” are in the range [7,9], for “Deceased/10000 inh.” in [7,9] and for “Natural increase/10000 inh.” in [-1,1].

Percentiles length indicators analysis (Figure 378) show that, indeed the concentration is around the middle of the data.

Figure 378

A comparison of the indicator “Live births” with the national level shows that it is better than the national, being better in 78.13% cases. For “Deceased” the indicator is better than the national, being better in 97.92% cases. Finally, for “Natural increase”, the indicator is better than the national, being better in 98.96% cases.

Figure 379

Regression analysis relative to indicator “Marriages” gives us an equation: y=-0.937310092x+275.9699561 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Divorces” gives us an equation: y=-0.215809821x+67.47719298 where x is the number of month (Jan, 2007=1), therefore a downward trend.

For the set of values above, the median indicator for “Marriages” is 216 and for “Divorces” is 53. Also, the distribution of quartiles is for “Marriages”: (51,111.5,216,325,593) and for “Divorces”: (11,36,52.5,74.25,169). The arithmetic mean and the standard deviation for “Marriages” are: (231,134.13) and for “Divorces”: (57,29.05). This means that with a probability greather than 0.68 “Marriages” are in the range [97,365] and for “Divorces” in [28,86].

Percentiles length indicators analysis (Figure 380) show that, indeed the concentration is around the middle of the data.

Figure 380

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Marriages/10000 inh. and Divorces/10000 inh. as in the figure 381.

Figure 381

Regression analysis relative to indicator “Marriages/10000 inh.” gives us an equation: y=-0.02142207x+6.050324561 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Divorces/10000 inh.” gives us an equation: y=-0.004904097x+1.476598684 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Marriages/10000 inh.” is 5 and for “Divorces/10000 inh.” is 1. Also, the distribution of quartiles is for “Marriages/10000 inh.”: (1.1,2.4175,4.695,7.1275,12.96) and for “Divorces/10000 inh.”: (0.24,0.7875,1.15,1.615,3.66). The arithmetic mean and the standard deviation for “Marriages/10000 inh.” are: (5,2.92) and for “Divorces/10000 inh.”: (1,0.63). This means that with a probability greather than 0.68 “Marriages/10000 inh.” are in the range [2,8] and for “Divorces/10000 inh.” in [0,2].

Percentiles length indicators analysis (Figure 382) show that, indeed the concentration is around the middle of the data.

Figure 382

A comparison of the indicator “Marriages” with the national level shows that it is better than the national, being better in 68.75% cases. For “Divorces” the indicator is about the same with the national, being better in 54.17% cases.

Figure 383

Regression analysis relative to indicator “Deaths under 1 year” gives us an equation: y=-0.024403147x+4.537719298 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year” is 3 and the distribution of quartiles is for “Deaths under 1 year”: (0,2,3,4.25,10). The arithmetic mean and the standard deviation for “Deaths under 1 year” are: (3,2.02) which means that with a probability greather than 0.68 “Deaths under 1 year” are in the range [1,5]. Percentiles length indicators analysis (Figure 384) show that, indeed the concentration is around the middle of the data.

Figure 384

Figure 385

Regression analysis relative to indicator “Deaths under 1 year/100000 inh.” gives us an equation: y=-0.005472599x+0.995004386 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year/100000 inh.” is 1 and the distribution of quartiles is for “Deaths under 1 year/100000 inh.”: (0,0.43,0.65,0.93,2.19). The arithmetic mean and the standard deviation for “Deaths under 1 year/100000 inh.” are: (1,0.44) which means that with a probability greather than 0.68 “Deaths under 1 year/100000 inh.” are in the range [1,1]. A comparison of the indicator “Deaths under 1 year” with the national level shows that it is better than the national, being better in 60.42% cases. A final analysis examines dependence aforementioned indicators of regional GDP variation.

Table 210. The evolution of Sibiu County GDP during 2007-2014

Source: INSSE and own calculations

In what follows, we shall investigate if there is a dependency between GDP variation (noted with dGDP) and the aforementioned indicators.

Searching dependence annual variations of “Live births” from GDP, we find that there is a dependence of Live births from GDP in the current year and the regression equation is: 0.8537dGDP+-0.0617. Searching dependence annual variations of “Deceased” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Natural increase” from GDP, we find that there is a dependence of Natural increase from GDP in the current year and the regression equation is: 119.7435dGDP+150.5133. Searching dependence annual variations of “Marriages” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Divorces” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Deaths under 1 year” from GDP, we find that there is a dependence of Deaths under 1 year from GDP offset by 2 years and the regression equation is:-3.6896dGDP+0.8227.

2.36. Analysis of Natural Movement of Suceava County Population

Statistics of natural movement corresponding to Suceava County are the following:

Table 211. The natural movement of Suceava County population during 2007-2008

Source: INSSE

Table 212. The natural movement of Suceava County population during 2009-2010

Source: INSSE

Table 213. The natural movement of Suceava County population during 2011-2012

Source: INSSE

Table 214. The natural movement of Suceava County population during 2013-2014

Source: INSSE

Table 215. The population trends of Suceava County during 2007-2014

Source: INSSE

Figure 386

From figure 386 we can see a sinusoidal evolution of the indicator. Except months feb 2007, mar 2007, apr 2007, mai 2007, iun 2007, iul 2007, aug 2007, sept 2007, oct 2007, ian 2008, feb 2008, mar 2008, apr 2008, mai 2008, iun 2008, iul 2008, aug 2008, sept 2008, oct 2008, nov 2008, apr 2009, mai 2009, iun 2009, iul 2009, aug 2009, sept 2009, oct 2009, nov 2009, mar 2010, mai 2010, iun 2010, iul 2010, aug 2010, sept 2010, oct 2010, iun 2011, iul 2011, aug 2011, sept 2011, oct 2011, mai 2012, iun 2012, iul 2012, aug 2012, sept 2012, oct 2012, ian 2013, mai 2013, iun 2013, iul 2013, aug 2013, sept 2013, oct 2013, nov 2013, mai 2014, iun 2014, iul 2014, aug 2014, sept 2014, oct 2014 the natural increase was negative.

Regression analysis relative to indicator “Live births” gives us an equation: y=-0.334393652x+695.8326754 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Deceased” gives us an equation: y=0.149559143x+619.7151316 where x is the number of month (Jan, 2007=1), therefore an upward trend.

Regression analysis relative to indicator “Natural increase” gives us an equation: y=-0.483952794x+76.11754386 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

For the set of values above, the median indicator for “Live births” is 663, for “Deceased” is 625 and for “Natural increase”: 32. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this. Also, the distribution of quartiles is for “Live births”: (517,598.75,662.5,727.75,1093), for “Deceased”: (536,586.25,624.5,663.5,791) and for “Natural increase”: (-208,-45.25,31.5,137.75,525). The arithmetic mean and the standard deviation for “Live births” are: (680,113.28), for “Deceased”: (627,54.64) and for “Natural increase”: (53,148.19). This means that with a probability greather than 0.68 “Live births” are in the range [567,793], for “Deceased” in [572,682] and for “Natural increase” in [-95,201].

Percentiles length indicators analysis (Figure 387) show that, indeed the concentration is around the middle of the data.

Figure 387

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Live births/10000 inh., Deceased/10000 inh. and Natural increase/10000 inh. as in the figure 388.

Figure 388

Regression analysis relative to indicator “Live births/10000 inh.” gives us an equation: y=-0.005761937x+9.500912281 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Deceased/10000 inh.” gives us an equation: y=0.000924783x+8.460252193 where x is the number of month (Jan, 2007=1), therefore a very small upward trend.

Regression analysis relative to indicator “Natural increase/10000 inh.” gives us an equation: y=-0.006667661x+1.037964912 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Live births/10000 inh.” is 9, for “Deceased/10000 inh.” is 8 and for “Natural increase/10000 inh.”: 0. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births/10000 inh.”: (7,8.1175,8.98,9.8875,14.82), for “Deceased/10000 inh.”: (7.24,7.965,8.46,9.015,10.79) and for “Natural increase/10000 inh.”: (-2.81,-0.615,0.43,1.8675,7.12).

The arithmetic mean and the standard deviation for “Live births/10000 inh.” are: (9,1.54), for “Deceased/10000 inh.”: (9,0.74) and for “Natural increase/10000 inh.”: (1,2.01). This means that with a probability greather than 0.68 “Live births/10000 inh.” are in the range [7,11], for “Deceased/10000 inh.” in [8,10] and for “Natural increase/10000 inh.” in [-1,3].

Percentiles length indicators analysis (Figure 389) show that, indeed the concentration is around the middle of the data.

Figure 389

A comparison of the indicator “Live births” with the national level shows that it is better than the national, being better in 100% cases. For “Deceased” the indicator is better than the national, being better in 97.92% cases. Finally, for “Natural increase”, the indicator is better than the national, being better in 100% cases.

Figure 390

Regression analysis relative to indicator “Marriages” gives us an equation: y=-0.711055345x+429.8403509 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Divorces” gives us an equation: y=-0.414744981x+112.1567982 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

For the set of values above, the median indicator for “Marriages” is 281 and for “Divorces” is 94. Also, the distribution of quartiles is for “Marriages”: (70,179,280.5,514.75,1450) and for “Divorces”: (1,66.75,94,117.25,161). The arithmetic mean and the standard deviation for “Marriages” are: (395,297.76) and for “Divorces”: (92,35.53). This means that with a probability greather than 0.68 “Marriages” are in the range [97,693] and for “Divorces” in [56,128].

Percentiles length indicators analysis (Figure 391) show that, indeed the concentration is around the middle of the data.

Figure 391

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Marriages/10000 inh. and Divorces/10000 inh. as in the figure 392.

Figure 392

Regression analysis relative to indicator “Marriages/10000 inh.” gives us an equation: y=-0.010410404x+5.870425439 where x is the number of month (Jan, 2007=1), therefore a very small downward trend. Regression analysis relative to indicator “Divorces/10000 inh.” gives us an equation: y=-0.005798359x+1.530699561 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Marriages/10000 inh.” is 4 and for “Divorces/10000 inh.” is 1. Also, the distribution of quartiles is for “Marriages/10000 inh.”: (0.95,2.4275,3.79,6.9475,19.75) and for “Divorces/10000 inh.”: (0.01,0.9075,1.275,1.59,2.18). The arithmetic mean and the standard deviation for “Marriages/10000 inh.” are: (5,4.04) and for “Divorces/10000 inh.”: (1,0.48). This means that with a probability greather than 0.68 “Marriages/10000 inh.” are in the range [1,9] and for “Divorces/10000 inh.” in [1,1].

Percentiles length indicators analysis (Figure 393) show that, indeed the concentration is around the middle of the data.

Figure 393

A comparison of the indicator “Marriages” with the national level shows that it is about the same with the national, being better in 55.21% cases. For “Divorces” the indicator is worse than the national, being better only in 39.58% cases.

Figure 394

Regression analysis relative to indicator “Deaths under 1 year” gives us an equation: y=-0.037452523x+8.608114035 where x is the number of month (Jan, 2007=1), therefore a very small downward trend. For the set of values above, the median indicator for “Deaths under 1 year” is 6 and the distribution of quartiles is for “Deaths under 1 year”: (3,5,6,8,15). The arithmetic mean and the standard deviation for “Deaths under 1 year” are: (7,2.67) which means that with a probability greather than 0.68 “Deaths under 1 year” are in the range [4,10].

Percentiles length indicators analysis (Figure 395) show that, indeed the concentration is around the middle of the data.

Figure 395

Figure 396

Regression analysis relative to indicator “Deaths under 1 year/100000 inh.” gives us an equation: y=-0.005193706x+1.173561404 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year/100000 inh.” is 1 and the distribution of quartiles is for “Deaths under 1 year/100000 inh.”: (0.4,0.68,0.815,1.09,2.05). The arithmetic mean and the standard deviation for “Deaths under 1 year/100000 inh.” are: (1,0.36) which means that with a probability greather than 0.68 “Deaths under 1 year/100000 inh.” are in the range [1,1].

A comparison of the indicator “Deaths under 1 year” with the national level shows that it is worse than the national, being better only in 37.5% cases.

A final analysis examines dependence aforementioned indicators of regional GDP variation.

Table 216. The evolution of Suceava County GDP during 2007-2014

Source: INSSE and own calculations

In what follows, we shall investigate if there is a dependency between GDP variation (noted with dGDP) and the aforementioned indicators.

Searching dependence annual variations of “Live births” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Deceased” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Natural increase” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Marriages” from GDP, we find that there is a dependence of Marriages from GDP offset by 1 year and the regression equation is:1.8462dGDP+1.1026. Searching dependence annual variations of “Divorces” from GDP, we find that there is a dependence of Divorces from GDP offset by 2 years and the regression equation is:-3.9728dGDP+-11.9191. Searching dependence annual variations of “Deaths under 1 year” from GDP, we find that there is not a dependence of the variation of GDP.

2.37. Analysis of Natural Movement of Teleorman County Population

Statistics of natural movement corresponding to Teleorman County are the following:

Table 217. The natural movement of Teleorman County population during 2007-2008

Source: INSSE

Table 218. The natural movement of Teleorman County population during 2009-2010

Source: INSSE

Table 219. The natural movement of Teleorman County population during 2011-2012

Source: INSSE

Table 220. The natural movement of Teleorman County population during 2013-2014

Source: INSSE

Table 221. The population trends of Teleorman County during 2007-2014

Source: INSSE

Figure 397

From figure 397 we can see a sinusoidal evolution of the indicator. #VALUE!

Regression analysis relative to indicator “Live births” gives us an equation: y=-0.314256647x+265.158114 where x is the number of month (Jan, 2007=1), therefore a downward trend.

Regression analysis relative to indicator “Deceased” gives us an equation: y=-0.394682583x+587.1212719 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Natural increase” gives us an equation: y=0.080425936x+-321.9631579 where x is the number of month (Jan, 2007=1), therefore an upward trend.

For the set of values above, the median indicator for “Live births” is 252, for “Deceased” is 567 and for “Natural increase”: -321. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births”: (166,227,251.5,273.25,372), for “Deceased”: (419,499.75,567,616,761) and for “Natural increase”: (-521,-377,-320.5,-245.5,-125).

The arithmetic mean and the standard deviation for “Live births” are: (250,35.62), for “Deceased”: (568,74.54) and for “Natural increase”: (-318,88.73). This means that with a probability greather than 0.68 “Live births” are in the range [214,286], for “Deceased” in [493,643] and for “Natural increase” in [-407,-229].

Percentiles length indicators analysis (Figure 398) show that, indeed the concentration is around the middle of the data.

Figure 398

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Live births/10000 inh., Deceased/10000 inh. and Natural increase/10000 inh. as in the figure 399.

Figure 399

Regression analysis relative to indicator “Live births/10000 inh.” gives us an equation: y=-0.002135988x+6.107657895 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Deceased/10000 inh.” gives us an equation: y=0.002812873x+13.51680482 where x is the number of month (Jan, 2007=1), therefore a very small upward trend.

Regression analysis relative to indicator “Natural increase/10000 inh.” gives us an equation: y=-0.004956932x+-7.409276316 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Live births/10000 inh.” is 6, for “Deceased/10000 inh.” is 13 and for “Natural increase/10000 inh.”: -8. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births/10000 inh.”: (4.1,5.4175,6.035,6.5675,8.79), for “Deceased/10000 inh.”: (9.71,12.0575,13.47,14.8475,17.8) and for “Natural increase/10000 inh.”: (-12.31,-9.18,-7.69,-5.8625,-2.95).

The arithmetic mean and the standard deviation for “Live births/10000 inh.” are: (6,0.83), for “Deceased/10000 inh.”: (14,1.78) and for “Natural increase/10000 inh.”: (-8,2.14). This means that with a probability greather than 0.68 “Live births/10000 inh.” are in the range [5,7], for “Deceased/10000 inh.” in [12,16] and for “Natural increase/10000 inh.” in [-10,-6].

Percentiles length indicators analysis (Figure 400) show that, indeed the concentration is around the middle of the data.

Figure 400

A comparison of the indicator “Live births” with the national level shows that it is worse than the national, being better only in 2.08% cases. For “Deceased” the indicator is worse than the national, being better only in 0% cases. Finally, for “Natural increase”, the indicator is worse than the national, being better only in 0% cases.

Figure 401

Regression analysis relative to indicator “Marriages” gives us an equation: y=-1.827061856x+241.4666667 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Divorces” gives us an equation: y=-0.197307379x+53.39232456 where x is the number of month (Jan, 2007=1), therefore a downward trend.

For the set of values above, the median indicator for “Marriages” is 131 and for “Divorces” is 42. Also, the distribution of quartiles is for “Marriages”: (25,61.5,130.5,228,602) and for “Divorces”: (12,31.75,42,54,91). The arithmetic mean and the standard deviation for “Marriages” are: (153,108.38) and for “Divorces”: (44,16.8). This means that with a probability greather than 0.68 “Marriages” are in the range [45,261] and for “Divorces” in [27,61].

Percentiles length indicators analysis (Figure 402) show that, indeed the concentration is around the middle of the data.

Figure 402

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Marriages/10000 inh. and Divorces/10000 inh. as in the figure 403.

Figure 403

Regression analysis relative to indicator “Marriages/10000 inh.” gives us an equation: y=-0.039948182x+5.582486842 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Divorces/10000 inh.” gives us an equation: y=-0.003795985x+1.23452193 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Marriages/10000 inh.” is 3 and for “Divorces/10000 inh.” is 1. Also, the distribution of quartiles is for “Marriages/10000 inh.”: (0.63,1.485,3.1,5.4025,13.95) and for “Divorces/10000 inh.”: (0.29,0.7575,1.005,1.29,2.13). The arithmetic mean and the standard deviation for “Marriages/10000 inh.” are: (4,2.53) and for “Divorces/10000 inh.”: (1,0.39). This means that with a probability greather than 0.68 “Marriages/10000 inh.” are in the range [1,7] and for “Divorces/10000 inh.” in [1,1].

Percentiles length indicators analysis (Figure 404) show that, indeed the concentration is around the middle of the data.

Figure 404

A comparison of the indicator “Marriages” with the national level shows that it is worse than the national, being better only in 7.29% cases. For “Divorces” the indicator is about the same with the national, being better in 59.38% cases.

Figure 405

Regression analysis relative to indicator “Deaths under 1 year” gives us an equation: y=-0.018319316x+3.836403509 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year” is 3 and the distribution of quartiles is for “Deaths under 1 year”: (0,1.75,3,4,8). The arithmetic mean and the standard deviation for “Deaths under 1 year” are: (3,1.84) which means that with a probability greather than 0.68 “Deaths under 1 year” are in the range [1,5].

Percentiles length indicators analysis (Figure 406) show that, indeed the concentration is around the middle of the data.

Figure 406

Figure 407

Regression analysis relative to indicator “Deaths under 1 year/100000 inh.” gives us an equation: y=-0.003743625x+0.886982456 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year/100000 inh.” is 1 and the distribution of quartiles is for “Deaths under 1 year/100000 inh.”: (0,0.4075,0.72,0.95,1.91). The arithmetic mean and the standard deviation for “Deaths under 1 year/100000 inh.” are: (1,0.43) which means that with a probability greather than 0.68 “Deaths under 1 year/100000 inh.” are in the range [1,1].

A comparison of the indicator “Deaths under 1 year” with the national level shows that it is about the same with the national, being better in 55.21% cases.

A final analysis examines dependence aforementioned indicators of regional GDP variation.

Table 222. The evolution of Teleorman County GDP during 2007-2014

Source: INSSE and own calculations

In what follows, we shall investigate if there is a dependency between GDP variation (noted with dGDP) and the aforementioned indicators.

Searching dependence annual variations of “Live births” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Deceased” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Natural increase” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Marriages” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Divorces” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Deaths under 1 year” from GDP, we find that there is not a dependence of the variation of GDP.

2.38. Analysis of Natural Movement of Timis County Population

Statistics of natural movement corresponding to Timis County are the following:

Table 223. The natural movement of Timis County population during 2007-2008

Source: INSSE

Table 224. The Natural Movement of Timis County Population during 2009-2010

Source: INSSE

Table 225. The natural movement of Timis County population during 2011-2012

Source: INSSE

Table 226. The natural movement of Timis County population during 2013-2014

Source: INSSE

Table 227. The population trends of Timis County during 2007-2014

Source: INSSE

Figure 408

From figure 408 we can see a sinusoidal evolution of the indicator. Except months aug 2007, sept 2007, feb 2008, iul 2008, aug 2008, sept 2008, oct 2008, iun 2009, iul 2009, aug 2009, sept 2009, iun 2010, iul 2010, aug 2010, sept 2010, iul 2011, aug 2011, sept 2011, mai 2012, aug 2012, sept 2012, oct 2012, ian 2013, iul 2013, aug 2013, sept 2013, oct 2013, mai 2014, iul 2014, aug 2014, sept 2014, dec 201 the natural increase was negative.

Regression analysis relative to indicator “Live births” gives us an equation: y=-0.164073521x+568.7596491 where x is the number of month (Jan, 2007=1), therefore a downward trend.

Regression analysis relative to indicator “Deceased” gives us an equation: y=-0.515911557x+647.8967105 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Natural increase” gives us an equation: y=0.351838036x+-79.1370614 where x is the number of month (Jan, 2007=1), therefore a pronounced upward trend.

For the set of values above, the median indicator for “Live births” is 559, for “Deceased” is 602 and for “Natural increase”: -55. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births”: (436,511.5,558.5,610.25,688), for “Deceased”: (101,564.75,601.5,661,984) and for “Natural increase”: (-548,-128,-55,29.25,419).

The arithmetic mean and the standard deviation for “Live births” are: (561,60.92), for “Deceased”: (623,112.34) and for “Natural increase”: (-62,139.14). This means that with a probability greather than 0.68 “Live births” are in the range [500,622], for “Deceased” in [511,735] and for “Natural increase” in [-201,77].

Percentiles length indicators analysis (Figure 409) show that, indeed the concentration is around the middle of the data.

Figure 409

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Live births/10000 inh., Deceased/10000 inh. and Natural increase/10000 inh. as in the figure 410.

Figure 410

Regression analysis relative to indicator “Live births/10000 inh.” gives us an equation: y=-0.004889718x+7.939234649 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Deceased/10000 inh.” gives us an equation: y=-0.009930277x+9.037660088 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Natural increase/10000 inh.” gives us an equation: y=0.005062398x+-1.099276316 where x is the number of month (Jan, 2007=1), therefore an upward trend.

For the set of values above, the median indicator for “Live births/10000 inh.” is 8, for “Deceased/10000 inh.” is 8 and for “Natural increase/10000 inh.”: -1. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births/10000 inh.”: (5.95,7.05,7.75,8.4225,9.46), for “Deceased/10000 inh.”: (1.37,7.7325,8.285,9.0975,13.42) and for “Natural increase/10000 inh.”: (-7.48,-1.7775,-0.76,0.405,5.68).

The arithmetic mean and the standard deviation for “Live births/10000 inh.” are: (8,0.84), for “Deceased/10000 inh.”: (9,1.55) and for “Natural increase/10000 inh.”: (-1,1.9). This means that with a probability greather than 0.68 “Live births/10000 inh.” are in the range [7,9], for “Deceased/10000 inh.” in [7,11] and for “Natural increase/10000 inh.” in [-3,1].

Percentiles length indicators analysis (Figure 411) show that, indeed the concentration is around the middle of the data.

Figure 411

A comparison of the indicator “Live births” with the national level shows that it is about the same with the national, being better in 46.88% cases. For “Deceased” the indicator is better than the national, being better in 86.46% cases. Finally, for “Natural increase”, the indicator is better than the national, being better in 82.29% cases.

Figure 412

Regression analysis relative to indicator “Marriages” gives us an equation: y=-0.348629951x+404.6377193 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Divorces” gives us an equation: y=-0.117681769x+68.44714912 where x is the number of month (Jan, 2007=1), therefore a downward trend.

For the set of values above, the median indicator for “Marriages” is 401 and for “Divorces” is 59. Also, the distribution of quartiles is for “Marriages”: (95,238,400.5,500,904) and for “Divorces”: (6,37.75,59,74.75,398). The arithmetic mean and the standard deviation for “Marriages” are: (388,170.15) and for “Divorces”: (63,46.95). This means that with a probability greather than 0.68 “Marriages” are in the range [218,558] and for “Divorces” in [16,110].

Percentiles length indicators analysis (Figure 413) show that, indeed the concentration is around the middle of the data.

Figure 413

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Marriages/10000 inh. and Divorces/10000 inh. as in the figure 414.

Figure 414

Regression analysis relative to indicator “Marriages/10000 inh.” gives us an equation: y=-0.006720293x+5.653225877 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Divorces/10000 inh.” gives us an equation: y=-0.001985825x+0.9585 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Marriages/10000 inh.” is 6 and for “Divorces/10000 inh.” is 1. Also, the distribution of quartiles is for “Marriages/10000 inh.”: (1.3,3.2675,5.515,6.925,12.54) and for “Divorces/10000 inh.”: (0.08,0.515,0.805,1.02,5.56). The arithmetic mean and the standard deviation for “Marriages/10000 inh.” are: (5,2.35) and for “Divorces/10000 inh.”: (1,0.65). This means that with a probability greather than 0.68 “Marriages/10000 inh.” are in the range [3,7] and for “Divorces/10000 inh.” in [0,2]. Percentiles length indicators analysis (Figure 415) show that, indeed the concentration is around the middle of the data.

Figure 415

A comparison of the indicator “Marriages” with the national level shows that it is better than the national, being better in 73.96% cases. For “Divorces” the indicator is better than the national, being better in 82.29% cases.

Figure 416

Regression analysis relative to indicator “Deaths under 1 year” gives us an equation: y=-0.033288117x+6.926973684 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year” is 5 and the distribution of quartiles is for “Deaths under 1 year”: (1,3,5,7,14). The arithmetic mean and the standard deviation for “Deaths under 1 year” are: (5,2.79) which means that with a probability greather than 0.68 “Deaths under 1 year” are in the range [2,8].

Percentiles length indicators analysis (Figure 417) show that, indeed the concentration is around the middle of the data.

Figure 417

Figure 418

Regression analysis relative to indicator “Deaths under 1 year/100000 inh.” gives us an equation: y=-0.00482637x+0.965328947 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year/100000 inh.” is 1 and the distribution of quartiles is for “Deaths under 1 year/100000 inh.”: (0.14,0.42,0.685,0.96,1.91). The arithmetic mean and the standard deviation for “Deaths under 1 year/100000 inh.” are: (1,0.38) which means that with a probability greather than 0.68 “Deaths under 1 year/100000 inh.” are in the range [1,1].

A comparison of the indicator “Deaths under 1 year” with the national level shows that it is about the same with the national, being better in 58.33% cases.

A final analysis examines dependence aforementioned indicators of regional GDP variation.

Table 228.The evolution of Timis County GDP during 2007-2014

Source: INSSE and own calculations

In what follows, we shall investigate if there is a dependency between GDP variation (noted with dGDP) and the aforementioned indicators.

Searching dependence annual variations of “Live births” from GDP, we find that there is a dependence of Live births from GDP in the current year and the regression equation is: 0.3883dGDP+0.2659. Searching dependence annual variations of “Deceased” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Natural increase” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Marriages” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Divorces” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Deaths under 1 year” from GDP, we find that there is a dependence of Deaths under 1 year from GDP in the current year and the regression equation is: -1.8734dGDP+-1.066.

2.39. Analysis of Natural Movement of Tulcea County Population

Statistics of natural movement corresponding to Tulcea County are the following:

Table 229. The natural movement of Tulcea County population during 2007-2008

Source: INSSE

Table 230. The natural movement of Tulcea County population during 2009-2010

Source: INSSE

Table 231. The natural movement of Tulcea County population during 2011-2012

Source: INSSE

Table 232. The natural movement of Tulcea County population during 2013-2014

Source: INSSE

Table 233. The population trends of Tulcea County during 2007-2014

Source: INSSE

Figure 419

From figure 419 we can see a sinusoidal evolution of the indicator. Except months aug 2007, sept 2007, iul 2008, sept 2008, aug 2009, dec 2011, aug 2013, aug 2014 the natural increase was negative.

Regression analysis relative to indicator “Live births” gives us an equation: y=-0.47474905x+202.0982456 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Deceased” gives us an equation: y=0.058790016x+246.7528509 where x is the number of month (Jan, 2007=1), therefore an upward trend.

Regression analysis relative to indicator “Natural increase” gives us an equation: y=-0.48878866x+-43.7625 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

For the set of values above, the median indicator for “Live births” is 179, for “Deceased” is 250 and for “Natural increase”: -69. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births”: (124,161.5,178.5,198,257), for “Deceased”: (188,228,249.5,271.25,317) and for “Natural increase”: (-162,-101.5,-69,-42.25,150).

The arithmetic mean and the standard deviation for “Live births” are: (179,26.87), for “Deceased”: (250,29.37) and for “Natural increase”: (-67,49.54). This means that with a probability greather than 0.68 “Live births” are in the range [152,206], for “Deceased” in [221,279] and for “Natural increase” in [-117,-17].

Percentiles length indicators analysis (Figure 420) show that, indeed the concentration is around the middle of the data.

Figure 420

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Live births/10000 inh., Deceased/10000 inh. and Natural increase/10000 inh. as in the figure 421.

Figure 421

Regression analysis relative to indicator “Live births/10000 inh.” gives us an equation: y=-0.015349905x+7.800407895 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Deceased/10000 inh.” gives us an equation: y=0.006920578x+9.51070614 where x is the number of month (Jan, 2007=1), therefore an upward trend.

Regression analysis relative to indicator “Natural increase/10000 inh.” gives us an equation: y=-0.020483383x+-1.674993421 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Live births/10000 inh.” is 7, for “Deceased/10000 inh.” is 10 and for “Natural increase/10000 inh.”: -3. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births/10000 inh.”: (4.96,6.485,6.965,7.74,10), for “Deceased/10000 inh.”: (7.28,9.0125,9.915,10.7175,12.38) and for “Natural increase/10000 inh.”: (-6.44,-3.99,-2.695,-1.6775,5.93).

The arithmetic mean and the standard deviation for “Live births/10000 inh.” are: (7,1.02), for “Deceased/10000 inh.”: (10,1.17) and for “Natural increase/10000 inh.”: (-3,1.97). This means that with a probability greather than 0.68 “Live births/10000 inh.” are in the range [6,8], for “Deceased/10000 inh.” in [9,11] and for “Natural increase/10000 inh.” in [-5,-1].

Percentiles length indicators analysis (Figure 422) show that, indeed the concentration is around the middle of the data.

Figure 422

A comparison of the indicator “Live births” with the national level shows that it is worse than the national, being better only in 21.88% cases. For “Deceased” the indicator is worse than the national, being better only in 20.83% cases. Finally, for “Natural increase”, the indicator is worse than the national, being better only in 12.5% cases.

Figure 423

Regression analysis relative to indicator “Marriages” gives us an equation: y=-0.656667119x+138.7337719 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Divorces” gives us an equation: y=-0.227244981x+41.17763158 where x is the number of month (Jan, 2007=1), therefore a downward trend.

For the set of values above, the median indicator for “Marriages” is 87 and for “Divorces” is 28. Also, the distribution of quartiles is for “Marriages”: (26,45.75,86.5,142.5,350) and for “Divorces”: (1,19,28,40.25,97). The arithmetic mean and the standard deviation for “Marriages” are: (107,70.13) and for “Divorces”: (30,18.11). This means that with a probability greather than 0.68 “Marriages” are in the range [37,177] and for “Divorces” in [12,48].

Percentiles length indicators analysis (Figure 424) show that, indeed the concentration is around the middle of the data.

Figure 424

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Marriages/10000 inh. and Divorces/10000 inh. as in the figure 425.

Figure 425

Regression analysis relative to indicator “Marriages/10000 inh.” gives us an equation: y=-0.023724702x+5.355752193 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Divorces/10000 inh.” gives us an equation: y=-0.008431091x+1.594741228 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Marriages/10000 inh.” is 3 and for “Divorces/10000 inh.” is 1. Also, the distribution of quartiles is for “Marriages/10000 inh.”: (1.05,1.7925,3.44,5.6,13.61) and for “Divorces/10000 inh.”: (0.04,0.755,1.1,1.5675,3.76). The arithmetic mean and the standard deviation for “Marriages/10000 inh.” are: (4,2.74) and for “Divorces/10000 inh.”: (1,0.71). This means that with a probability greather than 0.68 “Marriages/10000 inh.” are in the range [1,7] and for “Divorces/10000 inh.” in [0,2].

Percentiles length indicators analysis (Figure 426) show that, indeed the concentration is around the middle of the data.

Figure 426

A comparison of the indicator “Marriages” with the national level shows that it is worse than the national, being better only in 22.92% cases. For “Divorces” the indicator is about the same with the national, being better in 52.08% cases.

Figure 427

Regression analysis relative to indicator “Deaths under 1 year” gives us an equation: y=-0.007250407x+2.549561404 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year” is 2 and the distribution of quartiles is for “Deaths under 1 year”: (0,1,2,3,6). The arithmetic mean and the standard deviation for “Deaths under 1 year” are: (2,1.39) which means that with a probability greather than 0.68 “Deaths under 1 year” are in the range [1,3]. Percentiles length indicators analysis (Figure 428) show that, indeed the concentration is around the middle of the data.

Figure 428

Figure 429

Regression analysis relative to indicator “Deaths under 1 year/100000 inh.” gives us an equation: y=-0.002444995x+0.984936404 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year/100000 inh.” is 1 and the distribution of quartiles is for “Deaths under 1 year/100000 inh.”: (0,0.4,0.8,1.19,2.33). The arithmetic mean and the standard deviation for “Deaths under 1 year/100000 inh.” are: (1,0.54) which means that with a probability greather than 0.68 “Deaths under 1 year/100000 inh.” are in the range [0,2].

A comparison of the indicator “Deaths under 1 year” with the national level shows that it is about the same with the national, being better in 44.79% cases.

A final analysis examines dependence aforementioned indicators of regional GDP variation.

Table 234. The evolution of Tulcea County GDP during 2007-2014

Source: INSSE and own calculations

In what follows, we shall investigate if there is a dependency between GDP variation (noted with dGDP) and the aforementioned indicators.

Searching dependence annual variations of “Live births” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Deceased” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Natural increase” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Marriages” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Divorces” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Deaths under 1 year” from GDP, we find that there is a dependence of Deaths under 1 year from GDP offset by 1 year and the regression equation is:-3.2779dGDP+10.0866.

2.40. Analysis of Natural Movement of Valcea County Population

Statistics of natural movement corresponding to Valcea County are the following:

Table 235. The natural movement of Valcea County population during 2007-2008

Source: INSSE

Table 236. The natural movement of Valcea County population during 2009-2010

Source: INSSE

Table 237. The natural movement of Valcea County population during 2011-2012

Source: INSSE

Table 238. The natural movement of Valcea County population during 2013-2014

Source: INSSE

Table 239. The population trends of Valcea County during 2007-2014

Source: INSSE

Figure 430

From figure 430 we can see a sinusoidal evolution of the indicator. Except months aug 2007, sept 2007, iul 2008, aug 2009, iul 2010, aug 2011, ian 2013 the natural increase was negative.

Regression analysis relative to indicator “Live births” gives us an equation: y=-0.630995659x+301.3116228 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Deceased” gives us an equation: y=-0.815111232x+408.6578947 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Natural increase” gives us an equation: y=0.184115572x+-107.3462719 where x is the number of month (Jan, 2007=1), therefore an upward trend.

For the set of values above, the median indicator for “Live births” is 273, for “Deceased” is 371 and for “Natural increase”: -101. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births”: (189,242,273,298,362), for “Deceased”: (249,331,371,411,463) and for “Natural increase”: (-238,-142.25,-100.5,-48,80).

The arithmetic mean and the standard deviation for “Live births” are: (271,36.64), for “Deceased”: (369,48.46) and for “Natural increase”: (-98,61.61). This means that with a probability greather than 0.68 “Live births” are in the range [234,308], for “Deceased” in [321,417] and for “Natural increase” in [-160,-36].

Percentiles length indicators analysis (Figure 431) show that, indeed the concentration is around the middle of the data.

Figure 431

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Live births/10000 inh., Deceased/10000 inh. and Natural increase/10000 inh. as in the figure 432.

Figure 432

Regression analysis relative to indicator “Live births/10000 inh.” gives us an equation: y=-0.013327591x+7.203679825 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Deceased/10000 inh.” gives us an equation: y=-0.017098481x+9.771567982 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Natural increase/10000 inh.” gives us an equation: y=0.003784794x+-2.568041667 where x is the number of month (Jan, 2007=1), therefore an upward trend.

For the set of values above, the median indicator for “Live births/10000 inh.” is 7, for “Deceased/10000 inh.” is 9 and for “Natural increase/10000 inh.”: -2. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births/10000 inh.”: (4.6,5.8925,6.63,7.1925,8.73), for “Deceased/10000 inh.”: (6.09,8.03,8.965,9.88,11.16) and for “Natural increase/10000 inh.”: (-5.82,-3.4425,-2.425,-1.1675,1.93).

The arithmetic mean and the standard deviation for “Live births/10000 inh.” are: (7,0.86), for “Deceased/10000 inh.”: (9,1.14) and for “Natural increase/10000 inh.”: (-2,1.49). This means that with a probability greather than 0.68 “Live births/10000 inh.” are in the range [6,8], for “Deceased/10000 inh.” in [8,10] and for “Natural increase/10000 inh.” in [-3,-1].

Percentiles length indicators analysis (Figure 433) show that, indeed the concentration is around the middle of the data.

Figure 433

A comparison of the indicator “Live births” with the national level shows that it is worse than the national, being better only in 4.17% cases. For “Deceased” the indicator is better than the national, being better in 62.5% cases. Finally, for “Natural increase”, the indicator is worse than the national, being better only in 23.96% cases.

Figure 434

Regression analysis relative to indicator “Marriages” gives us an equation: y=-1.391603364x+232.9302632 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Divorces” gives us an equation: y=-0.169214596x+45.25899123 where x is the number of month (Jan, 2007=1), therefore a downward trend.

For the set of values above, the median indicator for “Marriages” is 132 and for “Divorces” is 35. Also, the distribution of quartiles is for “Marriages”: (35,75.25,131.5,239,527) and for “Divorces”: (0,23.75,35,50,121). The arithmetic mean and the standard deviation for “Marriages” are: (165,112.99) and for “Divorces”: (37,21.4). This means that with a probability greather than 0.68 “Marriages” are in the range [52,278] and for “Divorces” in [16,58].

Percentiles length indicators analysis (Figure 435) show that, indeed the concentration is around the middle of the data.

Figure 435

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Marriages/10000 inh. and Divorces/10000 inh. as in the figure 436.

Figure 436

Regression analysis relative to indicator “Marriages/10000 inh.” gives us an equation: y=-0.032439908x+5.574585526 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Divorces/10000 inh.” gives us an equation: y=-0.003800461x+1.081614035 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Marriages/10000 inh.” is 3 and for “Divorces/10000 inh.” is 1. Also, the distribution of quartiles is for “Marriages/10000 inh.”: (0.85,1.825,3.205,5.8125,12.66) and for “Divorces/10000 inh.”: (0,0.575,0.84,1.215,2.91). The arithmetic mean and the standard deviation for “Marriages/10000 inh.” are: (4,2.72) and for “Divorces/10000 inh.”: (1,0.52). This means that with a probability greather than 0.68 “Marriages/10000 inh.” are in the range [1,7] and for “Divorces/10000 inh.” in [0,2].

Percentiles length indicators analysis (Figure 437) show that, indeed the concentration is around the middle of the data.

Figure 437

A comparison of the indicator “Marriages” with the national level shows that it is worse than the national, being better only in 8.33% cases. For “Divorces” the indicator is better than the national, being better in 75% cases.

Figure 438

Regression analysis relative to indicator “Deaths under 1 year” gives us an equation: y=-0.007589528x+3.003508772 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year” is 2 and the distribution of quartiles is for “Deaths under 1 year”: (0,2,2,3.25,8). The arithmetic mean and the standard deviation for “Deaths under 1 year” are: (3,1.57) which means that with a probability greather than 0.68 “Deaths under 1 year” are in the range [1,5].

Percentiles length indicators analysis (Figure 439) show that, indeed the concentration is around the middle of the data.

Figure 439

Figure 440

Regression analysis relative to indicator “Deaths under 1 year/100000 inh.” gives us an equation: y=-0.001666848x+0.718758772 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year/100000 inh.” is 0 and the distribution of quartiles is for “Deaths under 1 year/100000 inh.”: (0,0.48,0.49,0.795,1.93). The arithmetic mean and the standard deviation for “Deaths under 1 year/100000 inh.” are: (1,0.38) which means that with a probability greather than 0.68 “Deaths under 1 year/100000 inh.” are in the range [1,1].

A comparison of the indicator “Deaths under 1 year” with the national level shows that it is better than the national, being better in 67.71% cases.

A final analysis examines dependence aforementioned indicators of regional GDP variation.

Table 240. The evolution of Valcea County GDP during 2007-2014

Source: INSSE and own calculations

In what follows, we shall investigate if there is a dependency between GDP variation (noted with dGDP) and the aforementioned indicators. Searching dependence annual variations of “Live births” from GDP, we find that there is a dependence of Live births from GDP offset by 2 years and the regression equation is:0.5842dGDP+-2.0983. Searching dependence annual variations of “Deceased” from GDP, we find that there is a dependence of Deceased from GDP in the current year and the regression equation is: -0.3901dGDP+-2.5867. Searching dependence annual variations of “Natural increase” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Marriages” from GDP, we find that there is a dependence of Marriages from GDP offset by 1 year and the regression equation is:0.7518dGDP+-7.4842. Searching dependence annual variations of “Divorces” from GDP, we find that there is a dependence of Divorces from GDP offset by 2 years and the regression equation is:-6.316dGDP+-0.7584. Searching dependence annual variations of “Deaths under 1 year” from GDP, we find that there is a dependence of Deaths under 1 year from GDP offset by 1 year and the regression equation is:-4.79dGDP+-1.7676.

2.41. Analysis of Natural Movement of Vaslui County Population

Statistics of natural movement corresponding to Vaslui County are the following:

Table 241. The natural movement of Vaslui County population during 2007-2008

Source: INSSE

Table 242. The natural movement of Vaslui County population during 2009-2010

Source: INSSE

Table 243. The natural movement of Vaslui County population during 2011-2012

Source: INSSE

Table 244. The natural movement of Vaslui County population during 2013-2014

Source: INSSE

Table 245. The population trends of Vaslui County during 2007-2014

Source: INSSE

Figure 441

From figure 441 we can see a sinusoidal evolution of the indicator. Except months mar 2007, apr 2007, mai 2007, iun 2007, iul 2007, aug 2007, sept 2007, iul 2008, aug 2008, sept 2008, oct 2008, iun 2009, iul 2009, aug 2009, sept 2009, oct 2009, sept 2010, iul 2011, aug 2011, sept 2011, aug 2012, sept 2012, oct 2012, iul 2013, aug 2013, sept 2013, iul 2014, aug 2014, sept 2014 the natural increase was negative.

Regression analysis relative to indicator “Live births” gives us an equation: y=-1.04156267x+447.8282895 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Deceased” gives us an equation: y=0.035885784x+446.1449561 where x is the number of month (Jan, 2007=1), therefore an upward trend.

Regression analysis relative to indicator “Natural increase” gives us an equation: y=-1.077448454x+1.683333333 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

For the set of values above, the median indicator for “Live births” is 392, for “Deceased” is 447 and for “Natural increase”: -50. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births”: (297,343,391.5,440.25,576), for “Deceased”: (318,401,447,495,601) and for “Natural increase”: (-272,-129.5,-50,20.75,195).

The arithmetic mean and the standard deviation for “Live births” are: (397,62.42), for “Deceased”: (448,62.94) and for “Natural increase”: (-51,103.68). This means that with a probability greather than 0.68 “Live births” are in the range [335,459], for “Deceased” in [385,511] and for “Natural increase” in [-155,53].

Percentiles length indicators analysis (Figure 442) show that, indeed the concentration is around the middle of the data.

Figure 442

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Live births/10000 inh., Deceased/10000 inh. and Natural increase/10000 inh. as in the figure 443.

Figure 443

Regression analysis relative to indicator “Live births/10000 inh.” gives us an equation: y=-0.020724973x+9.431932018 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Deceased/10000 inh.” gives us an equation: y=0.002174783x+9.399210526 where x is the number of month (Jan, 2007=1), therefore a very small upward trend.

Regression analysis relative to indicator “Natural increase/10000 inh.” gives us an equation: y=-0.022913388x+0.032861842 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Live births/10000 inh.” is 8, for “Deceased/10000 inh.” is 10 and for “Natural increase/10000 inh.”: -1. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births/10000 inh.”: (6.34,7.2825,8.35,9.3825,12.19), for “Deceased/10000 inh.”: (6.81,8.5425,9.535,10.555,12.87) and for “Natural increase/10000 inh.”: (-5.83,-2.74,-1.06,0.4375,4.13).

The arithmetic mean and the standard deviation for “Live births/10000 inh.” are: (8,1.3), for “Deceased/10000 inh.”: (10,1.34) and for “Natural increase/10000 inh.”: (-1,2.21). This means that with a probability greather than 0.68 “Live births/10000 inh.” are in the range [7,9], for “Deceased/10000 inh.” in [9,11] and for “Natural increase/10000 inh.” in [-3,1].

Percentiles length indicators analysis (Figure 444) show that, indeed the concentration is around the middle of the data.

Figure 444

A comparison of the indicator “Live births” with the national level shows that it is better than the national, being better in 86.46% cases. For “Deceased” the indicator is about the same with the national, being better in 43.75% cases. Finally, for “Natural increase”, the indicator is better than the national, being better in 69.79% cases.

Figure 445

Regression analysis relative to indicator “Marriages” gives us an equation: y=-1.287974769x+264.7063596 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Divorces” gives us an equation: y=-0.311923494x+78.79495614 where x is the number of month (Jan, 2007=1), therefore a downward trend.

For the set of values above, the median indicator for “Marriages” is 159 and for “Divorces” is 65. Also, the distribution of quartiles is for “Marriages”: (45,113.5,158.5,257,588) and for “Divorces”: (18,46,65,79,104). The arithmetic mean and the standard deviation for “Marriages” are: (202,124.83) and for “Divorces”: (64,20.29). This means that with a probability greather than 0.68 “Marriages” are in the range [77,327] and for “Divorces” in [44,84].

Percentiles length indicators analysis (Figure 446) show that, indeed the concentration is around the middle of the data.

Figure 446

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Marriages/10000 inh. and Divorces/10000 inh. as in the figure 447.

Figure 447

Regression analysis relative to indicator “Marriages/10000 inh.” gives us an equation: y=-0.026468597x+5.568622807 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Divorces/10000 inh.” gives us an equation: y=-0.006396093x+1.660627193 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Marriages/10000 inh.” is 3 and for “Divorces/10000 inh.” is 1. Also, the distribution of quartiles is for “Marriages/10000 inh.”: (0.96,2.415,3.375,5.4325,12.35) and for “Divorces/10000 inh.”: (0.38,0.9875,1.37,1.66,2.19). The arithmetic mean and the standard deviation for “Marriages/10000 inh.” are: (4,2.63) and for “Divorces/10000 inh.”: (1,0.43). This means that with a probability greather than 0.68 “Marriages/10000 inh.” are in the range [1,7] and for “Divorces/10000 inh.” in [1,1].

Percentiles length indicators analysis (Figure 448) show that, indeed the concentration is around the middle of the data.

Figure 448

A comparison of the indicator “Marriages” with the national level shows that it is worse than the national, being better only in 34.38% cases. For “Divorces” the indicator is worse than the national, being better only in 23.96% cases.

Figure 449

Regression analysis relative to indicator “Deaths under 1 year” gives us an equation: y=-0.044173901x+7.027850877 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year” is 4 and the distribution of quartiles is for “Deaths under 1 year”: (0,3,4,6,14). The arithmetic mean and the standard deviation for “Deaths under 1 year” are: (5,2.92) which means that with a probability greather than 0.68 “Deaths under 1 year” are in the range [2,8].

Percentiles length indicators analysis (Figure 450) show that, indeed the concentration is around the middle of the data.

Figure 450

Figure 451

Regression analysis relative to indicator “Deaths under 1 year/100000 inh.” gives us an equation: y=-0.009176886x+1.480078947 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year/100000 inh.” is 1 and the distribution of quartiles is for “Deaths under 1 year/100000 inh.”: (0,0.63,0.855,1.28,2.96). The arithmetic mean and the standard deviation for “Deaths under 1 year/100000 inh.” are: (1,0.62) which means that with a probability greather than 0.68 “Deaths under 1 year/100000 inh.” are in the range [0,2].

A comparison of the indicator “Deaths under 1 year” with the national level shows that it is worse than the national, being better only in 39.58% cases.

A final analysis examines dependence aforementioned indicators of regional GDP variation.

Table 246. The evolution of Vaslui County GDP during 2007-2014

Source: INSSE and own calculations

In what follows, we shall investigate if there is a dependency between GDP variation (noted with dGDP) and the aforementioned indicators.

Searching dependence annual variations of “Live births” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Deceased” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Natural increase” from GDP, we find that there is a dependence of Natural increase from GDP offset by 2 years and the regression equation is:18.5675dGDP+61.2646. Searching dependence annual variations of “Marriages” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Divorces” from GDP, we find that there is a dependence of Divorces from GDP in the current year and the regression equation is: -0.6494dGDP+-4.5242. Searching dependence annual variations of “Deaths under 1 year” from GDP, we find that there is a dependence of Deaths under 1 year from GDP offset by 2 years and the regression equation is:-0.4308dGDP+-9.3605.

2.42. Analysis of natural movement of Vrancea County population

Statistics of natural movement corresponding to Vrancea County are the following:

Table 247. The natural movement of Vrancea County population during 2007-2008

Source: INSSE

Table 248. The natural movement of Vrancea County population during 2009-2010

Source: INSSE

Table 249. The natural movement of Vrancea County population during 2011-2012

Source: INSSE

Table 250. The natural movement of Vrancea County population during 2013-2014

Source: INSSE

Table 251. The population trends of Vrancea County during 2007-2014

Source: INSSE

Figure 452

From figure 452 we can see a sinusoidal evolution of the indicator. Except months iun 2007, iul 2007, aug 2007, sept 2007, iul 2008, aug 2008, sept 2008, iul 2009, aug 2009, sept 2009, iul 2010, aug 2010, sept 2010, aug 2011, aug 2012, sept 2012, iul 2013, aug 2013, iul 2014, aug 2014 the natural increase was negative.

Regression analysis relative to indicator “Live births” gives us an equation: y=-0.532704829x+342.3778509 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Deceased” gives us an equation: y=0.052916441x+384.4960526 where x is the number of month (Jan, 2007=1), therefore an upward trend.

Regression analysis relative to indicator “Natural increase” gives us an equation: y=-0.58562127x+-42.11820175 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

For the set of values above, the median indicator for “Live births” is 307, for “Deceased” is 384 and for “Natural increase”: -81. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births”: (196,271.75,307,347.25,555), for “Deceased”: (260,354.25,384,416,518) and for “Natural increase”: (-254,-140.25,-80.5,-30.75,235).

The arithmetic mean and the standard deviation for “Live births” are: (317,64.5), for “Deceased”: (387,45.46) and for “Natural increase”: (-71,92.38). This means that with a probability greather than 0.68 “Live births” are in the range [253,382], for “Deceased” in [342,432] and for “Natural increase” in [-163,21].

Percentiles length indicators analysis (Figure 453) show that, indeed the concentration is around the middle of the data.

Figure 453

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Live births/10000 inh., Deceased/10000 inh. and Natural increase/10000 inh. as in the figure 454.

Figure 454

Regression analysis relative to indicator “Live births/10000 inh.” gives us an equation: y=-0.012156403x+8.546877193 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Deceased/10000 inh.” gives us an equation: y=0.002811177x+9.595532895 where x is the number of month (Jan, 2007=1), therefore a very small upward trend.

Regression analysis relative to indicator “Natural increase/10000 inh.” gives us an equation: y=-0.015001356x+-1.047017544 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Live births/10000 inh.” is 8, for “Deceased/10000 inh.” is 10 and for “Natural increase/10000 inh.”: -2. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births/10000 inh.”: (4.94,6.8575,7.7,8.695,13.98), for “Deceased/10000 inh.”: (6.51,8.8725,9.695,10.535,12.97) and for “Natural increase/10000 inh.”: (-6.4,-3.5325,-2.025,-0.77,5.92).

The arithmetic mean and the standard deviation for “Live births/10000 inh.” are: (8,1.62), for “Deceased/10000 inh.”: (10,1.14) and for “Natural increase/10000 inh.”: (-2,2.33). This means that with a probability greather than 0.68 “Live births/10000 inh.” are in the range [6,10], for “Deceased/10000 inh.” in [9,11] and for “Natural increase/10000 inh.” in [-4,0].

Percentiles length indicators analysis (Figure 455) show that, indeed the concentration is around the middle of the data.

Figure 455

A comparison of the indicator “Live births” with the national level shows that it is about the same with the national, being better in 48.96% cases. For “Deceased” the indicator is worse than the national, being better only in 23.96% cases. Finally, for “Natural increase”, the indicator is worse than the national, being better only in 37.5% cases.

Figure 456

Regression analysis relative to indicator “Marriages” gives us an equation: y=-1.248799512x+232.4105263 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Divorces” gives us an equation: y=-0.390518177x+57.58596491 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

For the set of values above, the median indicator for “Marriages” is 114 and for “Divorces” is 38. Also, the distribution of quartiles is for “Marriages”: (34,72,113.5,210.5,774) and for “Divorces”: (0,21.75,37.5,57.25,114). The arithmetic mean and the standard deviation for “Marriages” are: (172,158.45) and for “Divorces”: (39,24.52). This means that with a probability greather than 0.68 “Marriages” are in the range [14,330] and for “Divorces” in [14,64].

Percentiles length indicators analysis (Figure 457) show that, indeed the concentration is around the middle of the data.

Figure 457

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Marriages/10000 inh. and Divorces/10000 inh. as in the figure 458.

Figure 458

Regression analysis relative to indicator “Marriages/10000 inh.” gives us an equation: y=-0.030705168x+5.806179825 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Divorces/10000 inh.” gives us an equation: y=-0.009662371x+1.439458333 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Marriages/10000 inh.” is 3 and for “Divorces/10000 inh.” is 1. Also, the distribution of quartiles is for “Marriages/10000 inh.”: (0.86,1.81,2.845,5.285,19.38) and for “Divorces/10000 inh.”: (0,0.545,0.94,1.435,2.85). The arithmetic mean and the standard deviation for “Marriages/10000 inh.” are: (4,3.98) and for “Divorces/10000 inh.”: (1,0.61). This means that with a probability greather than 0.68 “Marriages/10000 inh.” are in the range [0,8] and for “Divorces/10000 inh.” in [0,2].

Percentiles length indicators analysis (Figure 459) show that, indeed the concentration is around the middle of the data.

Figure 459

A comparison of the indicator “Marriages” with the national level shows that it is worse than the national, being better only in 16.67% cases. For “Divorces” the indicator is better than the national, being better in 64.58% cases.

Figure 460

Regression analysis relative to indicator “Deaths under 1 year” gives us an equation: y=-0.054347531x+6.812938596 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year” is 3 and the distribution of quartiles is for “Deaths under 1 year”: (1,2,3,4.25,60). The arithmetic mean and the standard deviation for “Deaths under 1 year” are: (4,6.04) which means that with a probability greather than 0.68 “Deaths under 1 year” are in the range [-2,10]. Percentiles length indicators analysis (Figure 461) show that, indeed the concentration is around the middle of the data.

Figure 461

Figure 462

Regression analysis relative to indicator “Deaths under 1 year/100000 inh.” gives us an equation: y=-0.013458492x+1.700861842 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year/100000 inh.” is 1 and the distribution of quartiles is for “Deaths under 1 year/100000 inh.”: (0.25,0.51,0.76,1.07,15.02). The arithmetic mean and the standard deviation for “Deaths under 1 year/100000 inh.” are: (1,1.51) which means that with a probability greather than 0.68 “Deaths under 1 year/100000 inh.” are in the range [-1,3].

A comparison of the indicator “Deaths under 1 year” with the national level shows that it is worse than the national, being better only in 39.58% cases.

A final analysis examines dependence aforementioned indicators of regional GDP variation.

Table 252. The evolution of Vrancea County GDP during 2007-2014

Source: INSSE and own calculations

In what follows, we shall investigate if there is a dependency between GDP variation (noted with dGDP) and the aforementioned indicators.

Searching dependence annual variations of “Live births” from GDP, we find that there is a dependence of Live births from GDP in the current year and the regression equation is: 0.3518dGDP+-2.0883. Searching dependence annual variations of “Deceased” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Natural increase” from GDP, we find that there is a dependence of Natural increase from GDP offset by 2 years and the regression equation is:2.1701dGDP+9.8021. Searching dependence annual variations of “Marriages” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Divorces” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Deaths under 1 year” from GDP, we find that there is not a dependence of the variation of GDP.

2.43. Analysis of Natural Movement of Romania County Population

Statistics of natural movement corresponding to Romania County are the following:

Table 253. The natural movement of Romania County population during 2007-2008

Source: INSSE

Table 254. The natural movement of Romania County population during 2009-2010

Source: INSSE

Table 255. The natural movement of Romania County population during 2011-2012

Source: INSSE

Table 256. The natural movement of Romania County population during 2013-2014

Source: INSSE

Table 257. The population trends of Romania County during 2007-2014

Source: INSSE

Figure 463

From figure 463 we can see a sinusoidal evolution of the indicator. Except months aug 2007, sept 2007, iul 2008, aug 2008, sept 2008, iul 2009, aug 2009, sept 2009, aug 2011, sept 2011, aug 2012, sept 2012, aug 2013, iul 2014, aug 2014, sept 2014 the natural increase was negative.

Regression analysis relative to indicator “Live births” gives us an equation: y=-26.73288795x+18555.20132 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Deceased” gives us an equation: y=-3.932569181x+21352.75044 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Natural increase” gives us an equation: y=-22.80031877x+-2797.549123 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

For the set of values above, the median indicator for “Live births” is 17329, for “Deceased” is 21183 and for “Natural increase”: -4290. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births”: (13486,15833.25,17328.5,18731.5,21456), for “Deceased”: (17099,19707.25,21183,22135.5,25578) and for “Natural increase”: (-9885,-6289,-4290,-1316.75,2749).

The arithmetic mean and the standard deviation for “Live births” are: (17259,1977.72), for “Deceased”: (21162,1848.16) and for “Natural increase”: (-3903,3158.48). This means that with a probability greather than 0.68 “Live births” are in the range [15281,19237], for “Deceased” in [19314,23010] and for “Natural increase” in [-7061,-745].

Percentiles length indicators analysis (Figure 464) show that, indeed the concentration is around the middle of the data.

Figure 464

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Live births/10000 inh., Deceased/10000 inh. and Natural increase/10000 inh. as in the figure 465.

Figure 465

Regression analysis relative to indicator “Live births/10000 inh.” gives us an equation: y=-0.01094296x+8.206462719 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Deceased/10000 inh.” gives us an equation: y=-0.000589867x+9.441421053 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Natural increase/10000 inh.” gives us an equation: y=-0.010357434x+-1.235164474 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Live births/10000 inh.” is 8, for “Deceased/10000 inh.” is 9 and for “Natural increase/10000 inh.”: -2. This means that the probability that the indicator has a value less than the median is equal to the probability that it has a higher value than this.

Also, the distribution of quartiles is for “Live births/10000 inh.”: (6.02,7.045,7.69,8.3125,9.52), for “Deceased/10000 inh.”: (7.62,8.74,9.44,9.8475,11.33) and for “Natural increase/10000 inh.”: (-4.41,-2.7925,-1.905,-0.5825,1.22).

The arithmetic mean and the standard deviation for “Live births/10000 inh.” are: (8,0.87), for “Deceased/10000 inh.”: (9,0.82) and for “Natural increase/10000 inh.”: (-2,1.41). This means that with a probability greather than 0.68 “Live births/10000 inh.” are in the range [7,9], for “Deceased/10000 inh.” in [8,10] and for “Natural increase/10000 inh.” in [-3,-1]. Percentiles length indicators analysis (Figure 466) show that, indeed the concentration is around the middle of the data.

Figure 466

Figure 467

Regression analysis relative to indicator “Marriages” gives us an equation: y=-58.73511259x+13523.14254 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

Regression analysis relative to indicator “Divorces” gives us an equation: y=-7.06289338x+2891.456579 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

For the set of values above, the median indicator for “Marriages” is 10085 and for “Divorces” is 2603. Also, the distribution of quartiles is for “Marriages”: (2961,4950.75,10084.5,14994.75,29151) and for “Divorces”: (924,2206,2602.5,3040.5,3598). The arithmetic mean and the standard deviation for “Marriages” are: (10674,6298.66) and for “Divorces”: (2549,602.44). This means that with a probability greather than 0.68 “Marriages” are in the range [4375,16973] and for “Divorces” in [1947,3151].

Percentiles length indicators analysis (Figure 468) show that, indeed the concentration is around the middle of the data.

Figure 468

Taking into account the population dynamics during the analyzed period we have the following evolution of the indicators: Marriages/10000 inh. and Divorces/10000 inh. as in the figure 469.

Figure 469

Regression analysis relative to indicator “Marriages/10000 inh.” gives us an equation: y=-0.025459984x+5.980017544 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

Regression analysis relative to indicator “Divorces/10000 inh.” gives us an equation: y=-0.002999254x+1.278692982 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Marriages/10000 inh.” is 5 and for “Divorces/10000 inh.” is 1. Also, the distribution of quartiles is for “Marriages/10000 inh.”: (1.32,2.2025,4.5,6.645,12.92) and for “Divorces/10000 inh.”: (0.41,0.98,1.155,1.345,1.6). The arithmetic mean and the standard deviation for “Marriages/10000 inh.” are: (5,2.8) and for “Divorces/10000 inh.”: (1,0.27). This means that with a probability greather than 0.68 “Marriages/10000 inh.” are in the range [2,8] and for “Divorces/10000 inh.” in [1,1].

Percentiles length indicators analysis (Figure 470) show that, indeed the concentration is around the middle of the data.

Figure 470

Figure 471

Regression analysis relative to indicator “Deaths under 1 year” gives us an equation: y=-0.986550461x+217.795614 where x is the number of month (Jan, 2007=1), therefore a pronounced downward trend.

For the set of values above, the median indicator for “Deaths under 1 year” is 167 and the distribution of quartiles is for “Deaths under 1 year”: (108,146.25,166.5,192.5,292). The arithmetic mean and the standard deviation for “Deaths under 1 year” are: (170,33.53) which means that with a probability greather than 0.68 “Deaths under 1 year” are in the range [136,204].

Percentiles length indicators analysis (Figure 472) show that, indeed the concentration is around the middle of the data.

Figure 472

Figure 473

Regression analysis relative to indicator “Deaths under 1 year/100000 inh.” gives us an equation: y=-0.0042949x+0.963927632 where x is the number of month (Jan, 2007=1), therefore a very small downward trend.

For the set of values above, the median indicator for “Deaths under 1 year/100000 inh.” is 1 and the distribution of quartiles is for “Deaths under 1 year/100000 inh.”: (0.48,0.6475,0.74,0.8525,1.29). The arithmetic mean and the standard deviation for “Deaths under 1 year/100000 inh.” are: (1,0.15) which means that with a probability greather than 0.68 “Deaths under 1 year/100000 inh.” are in the range [1,1].

A final analysis examines dependence aforementioned indicators of regional GDP variation.

Table 258. The evolution of Romania County GDP during 2007-2014

Source: INSSE and own calculations

In what follows, we shall investigate if there is a dependency between GDP variation (noted with dGDP) and the aforementioned indicators.

Searching dependence annual variations of “Live births” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Deceased” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Natural increase” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Marriages” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Divorces” from GDP, we find that there is not a dependence of the variation of GDP. Searching dependence annual variations of “Deaths under 1 year” from GDP, we find that there is not a dependence of the variation of GDP.

Bibliography

Ioan, Gina & Ioan, Cătălin Angelo (2017). Macroeconomics. Galati: Zigotto Publishers.

Ioan, Cătălin Angelo (2011). Mathematics. Galati: Zigotto Publishers.

Voineagu, Vergil; Mitrut, Constantin & Isaic-Maniu Alexandru (2003). Statistics. Bucharest: Ed. Universitara.

Refbacks

• There are currently no refbacks.