Physics of auroral phenomena : proceedings of the 35th Annual seminar, Apatity, 28 Februaru – 02 March, 2012 / [ed. board: A. G. Yahnin, A. A. Mochalov]. - Апатиты : Издательство Кольского научного центра РАН, 2012. - 187 с. : ил., табл.

R E, Mikhailov et al. have bom in the years of low SA have lifespan on the 8.4 years more than in the group where people have bom in the years of high SA. Table 1. Summary statistics of distribution of Wolf numbers (W), life expectancy (age) for the entire study period indicators N mean median Mode Freq. mode min max St. Dev. st. error W 967 55,4 45,9 9,6 29 1,4 190,2 43,70 1,40 Age mean 967 59,5 64 80 34 18 94 19,51 0,63 Age (gr 1) 557 63,1 68 74 26 18 94 18,74 0,79 Age (gr 2) 410 54,6 59 70 17 18 93 19,47 0,96 Over the entire study period, the dead in the first group (at low values of the Wolf numbers) were more on 36% than in the second one, mainly due to the contribution to the overall mortality rate of people with age over 60 years. At the same time, in the second group, the younger people aged 20-30 years have introduced the main contribution to the overall mortality (Figure 2). That is, in the first group the mortality dominates in old age, and the young people who died at the age of 20-30 years make a significant contribution to mortality in the second group. To confirm the null hypothesis, which consists in assuming that high SA in the period of conceiving and prenatal development has a negative impact on life expectancy, the significance of the differences between the distribution of ages on groups was assessed. To do this, were applied the criteria nonparametric statistics (Mann- Whitney and Wald-Wolfowitz criterion), allowing to compare the distribution of the parameters in two independent groups. Group 1 Group 2 Fig. 2. Frequency distribution of life in 1 (low SA) and 2 (high SA) groups Mann-Whitney test is a nonparametric alternative to the t-test for independent samples with abnormal distribution. In the analysis using the Mann-Whitney test, the indicators (age) in the two groups are combined, arranged based on ascending point value and ranked. Rank values are summed for each group and use them to calculate the values of the criterion U (for each sample), used to assess the significance of differences between the two samples. U value of the criterion reflects how large area of agreement between the two samples and the smaller the value of U, so the differences are more significant. The results of analysis of the differences between the two groups are shown in Table 2. Table 2. Evaluation of the significance of differences between the two groups, corresponding to the birth in years Mann-Whitney U p <0,05 number of observations sum of the ranks U P gr- 1 gr-2 gr-1 gr-2 557 410 301007,0 167989,0 83734,00 <0,001 The data in Table 2 show that there are significant (p <0.001) differences between distributions in age of dead patients of two groups. According to the second criterion (Criterion series of Wald-Wolfowitz) indicators (age) in the two groups are combined, line up by degree of their ascending and then count the number of cases with adjacent values of age, belonging to the same group (with low or high SA in the year of birth ). If there are no differences between the 180