Statistical modeling for unemployment rate using smoothing spline in semiparametric multivariable regression model with Bayesian approach

Unemployment is a labor problem facing many countries. High or low levels of unemployment rate can be influenced by many factors, including level of education, investment, economic growth, population density, regional minimum wage, and the number of large and medium industries. It is extremely difficult to estimate the unemployment rate as it is not linearly affected by all the factors. But a semiparametric multivariable regression can be well applied for modeling the unemployment rate. Semiparametric multivariable regression includes regression models which combine parametric and nonparametric models. The nonparametric components can be approached by smoothing spline function. In this paper, the smoothing spline function to semiparametric multivariable regression with Bayesian approach for the unemployment rate is developed. Generalized Maximum Likelihood (GML) method was simultaneously executed to obtain the optimal smoothing spline parameters. By using the method, we found that the unemployment rate model in East Java province (Indonesia) in 2011 possess good performances with mean square error close to zero.