Spatial Modeling of Unemployment Rate in Counties of Iran Based on Population and Housing Census Data

Unemployment is one of the most important issues in all countries around the world. An increase in the number of unemployed in any society will cause many problems. Due to the high importance of the issue of unemployment, in many researches, the correct recognition and deep understanding of the factors affecting unemployment to reduce it has been considered. The current research collects data related to the Population and Housing Census 2016 of the Statistical Center of Iran. In these data, the economically active population and the number of unemployed, aged 15 years old or above are categorized by gender and different levels of education in Iran counties. The aim pursued during this research is the spatial modeling of the number of unemployed in counties of Iran, based on gender and education as covariates. To achieve this goal, the Bayesian approach and a method called “integrated nested Laplace approximation” or INLA for short, have been used. After fitting the appropriate spatial model, the coefficients of covariates have been estimated. The obtained results show that women's unemployment is higher than men's, and unemployment has also increased with the increase of education level compared to illiteracy. Finally, the spatial effects of the counties have been estimated and the counties with the highest and the lowest risk of unemployment, as well as the counties that have no significant difference with the average unemployment of the country, have been identified.