ANALYSIS OF THE FACTORS INFLUENCING UNEMPLOYMENT DURATION IN TURKEY BY DURATION MODELS

Abstract

Cox regression is one of the most commonly used methods in the literature for duration analysis. The purpose of the study is to examine variables that are thought to influence the duration of unemployment using the Cox regression model. The study of the years 2012-2015 Income and Living Conditions Survey used in the data set is taken from the Turkey Statistical Institute. Stratified Cox regression and time dependent explanatory variable Cox regression analyzes were used where the proof of proportional hazards assumption, which is the only basic assumption of Cox regression analysis, is not provided. As a result of the analysis, the models obtained were compared with Akaike and Bayes Information Criteria and the most suitable model was selected. According to the results of the most suitable model obtained, the marital status of the individual, education level, age, registration status with Social Security Institution, financial deprivation, household size, monthly net income received during the reference period and unemployment rate were determined as important variables affecting the unemployment duration.