| Abstract [eng] |
The object of this masters final work is the statistical and demographical data of Statistics Lithuania about Lithuanian residents over the period 2002-2018. The aim of this study is to determine how negative factors (crime, suicide and unemployment) affects Lithuanian residents internal movement within the country. To achieve this aim the following objectives have been set. First of all, to collect necessary data and identify statistical dependence between negative factors and internal migration nationwide. Next step is to group the data of Lithuania‘s regions, identify statistical dependence between negative factors and movement in the groups and between them. Another objective is to apply the statistical hypothesis about multinomial distribution and determine whether changes of movement and negative factors in groups are statistically significant every 4 years. Having identified this, changes in groups have to be marked in the map of Lithuania. And finally, check if there is a reason to apply time series models to statistical data about movement and negative factors in groups. Comparative analysis of internal migration and negative factors indicated, that there is average dependence between internal migrants, unemployment and crime, but there is a big dependence between internal migrants (y) and suicide (x). The following linear regression was found: y=85352,806-25,888x. Having applied the statistical hypothesis about multinomial distribution parameters the final work found out, that in 2002-2006; 2010-2014 unemployment changes were statistically significant with probability of 0,99. In 2002-2006 changes of arrivals, general movement, crime and unemployment were statistically significant with probability of 0,9 as well as changes of unemployment in 2006-2010 and 2010-2014. Having marked changes of data in the map of Lithuania it was found out, that negative factors does influence the movement of Lithuanian residents within the country. When there is decrease of arrivals in certain regions – unemployment increases in those same regions. When crime increases – arrivals decreases. Increased number of suicide shows bad living conditions in certain regions, this leads to decreased number of arrivals. The number of people that leaves regions where they live is directly related – when negative factors decreases, amount of people that left decreases too. Having checked the assumptions to apply time series, it was identified that time series models can be applied to the data of I group‘s crime and unemployment, II group‘s suicide, III group‘s arrivals, departed, general movement, crime and unemployment. AR(4) models were found for that data. |
