Research on Social Stability Based on PCA Analysis and Logistic Regression

Abstract

This paper investigates a series of regime change movements named after colors that occurred in the CIS and Central Asian regions in the early 21st century, conducted in peaceful and non-violent manners. Firstly, this study collects nine highly representative warning indicators in three aspects: "politics, economy, education," and assigns names to the three main components as "political indicators, economic indicators, cultural indicators" through principal component analysis, thereby constructing an indicator system affecting social stability. Secondly, stepwise regression analysis is used to screen out the significantly influential warning indicators on whether society is turbulent and a logistic model for social turbulence warning is constructed. Finally, relevant suggestions are once again offered in the three aspects of "politics, economy, education," providing effective resolution strategies for policy makers. The study concludes that the warning model of turbulence rate is related to the unemployment rate and the proportion of account balance, so the relevant departments need to timely adjust the unemployment rate and the proportion of account balance to maintain the safety level of social turbulence rate with the greatest efficiency.