Mathematical methods to assess factors affecting population health in the Russian regions (panel analysis) |
Tuesday, 12 November 2013 | ||||||
E.V. Molchanova1, M.M. Kruchek2 Summary. Health is the basis for public wellbeing of the nation, its economic and social prosperity. Study of the social and economic, environmental and psychological factors that affect medical and demographic trends is a topical issue for the modern scientific community. Complexity of this problem is that it is interdisciplinary in nature and its solution requires joint efforts of experts in different fields of knowledge; application of methods of mathematical statistics is important as well. The aim of the work. The article identifies factors that have significant impact on the main health indicator–life expectancy. Methods. The study is based on the Rosstat data ("Regions of Russia", "Health care in Russia"). At the first phase of the analysis information on the constituent entities of the Russian Federation for 2005-2010 was presented in the form of a special system that includes the following blocks: medico-demographic indicators, social and demographic structure of the population, social and economic development, access to medical services, climatic and natural conditions, and social stress. At the second phase, proceeding from the preliminary analysis, the most significant characteristics were selected. Then they were divided into groups so that the intra-group correlation of factors was relatively big, but small between. Method of correlation pleiades is one of the methods allowing for such operation. Creation of correlation pleiades helped to distinguish two main groups of characteristics: socio-economic indicators and social stress factors. This was followed by the panel data regression analysis. Panel data consist of observations of the same economic units or objects (firms, regions, countries, etc.), that are implemented in successive periods of time. Thanks to special structure the panel data allow to develop more flexible and substantial models. Scientific literature distinguishes three regression models according to panel data: pooled model, fixed effect model, and random effect model. Results and Conclusions. Calculations outlined in the article found out that the fixed effect model helps to get a meaningful and valid simulation option, which can be used to evaluate the main demographic indicator - life expectancy in the Russian regions, depending on the socio-economic development, health care and social stress factor Scope of application. The study results help to identify areas of work to decrease social and economic burden of diseases and improve quality of life of the population. Keywords: medico-demographic situation; region; health; life expectancy; causes of death. References
Views: 26314
Write Comment
|
||||||
Last Updated ( Tuesday, 12 November 2013 ) |
< Prev | Next > |
---|