The use of descriptive statistics for the characteristics of the empirical samples macroeconomic indicators

Abstract

In the article, the statistical tests of the empirical distributions of economic indicators of the Russian Federation in 2011 (average monthly nominal accrued salary, the unemployment rate), the unemployment rate of the countries of the European Union in 2011, on conformity to their normal distribution law (distribution fitting) are made. On the basis of the calculation of the statistical criterion of the Kolmogorov-Smirnov, (K-S test) the distributions being not normal distributions is proved. Therefore, for their characteristics it is inappropriate to use the arithmetic mean as indicators of economic status and development. For the selected empirical distributions of the calculations of such indicators descriptive statistics, as the median, the lower and upper quartiles, interquartile range are conducted, the advantages of graphical representation of the data in the form of a chart «box and whisker» are shown. On the example of the two Russian districts of the Southern Federal district and the Volga Federal district, the comparison of the average monthly salary in 2010 with the help of nonparametric test Mann-Whitney U test is carried out, the difference in the interpretation of data compared with the use of simple averages is shown. The results of the performed calculations can be used for the analysis of economic indicators average monthly salary and the unemployment rate in 83 subjects of the Russian Federation in 2011, the unemployment rate in the EU countries in 2011, as well as the methodology of statistical analysis of the different empirical distribution economic indicators.