Document Type : Original Research Paper
Authors
Department of Economic Sciences, School of Economics and Accounting, Islamic Azad University, Central Tehran Branch, Iran
Abstract
The aim of the current research is to investigate the existence of selection bias in Iran's life insurance market using demographic, economic-social information and factors affecting the risk of individuals using household income and expenditure statistics. The process of this research is that first, using the logit model, the effect of independent variables such as age, gender, education, marital status, activity status, household dimensions and variables related to income and investment status and the person's vision of the future (having insurance) Retirement and accident insurance) and the future health of the person is checked on the possibility of applying for life insurance. In the second step, the dependent variable was estimated in three groups based on the method of paying life insurance premiums in insured and uninsured households, segmentation and ordinal logit model, but due to the violation of the assumption of parallel regression lines in this model, in the end, the generalized logit model was used and analyzed. The final effects were discussed at each stage. Examining the coefficients of this model shows that the variables of age and high-risk jobs indicate the presence of selection bias in the life insurance market, but the variables of education, smoking, sports and religious expenses, medical expenses and income work to reduce selection bias in the life insurance market.
Keywords
Letters to Editor
Send comment about this article