Bayesian Tobit Principal Component Regression with Application Fadel Hamid Hadi Alhusseini

Tobit is considered an important statistical modeling, as it has become common in numerous practical applications, such as econometrics, biological sciences, finance, and medicine. The Tobit regression model is special case of censored regression model at zero point. In some cases, Tobit regression model is exposed to econometrics problems, including the multicollinearity problem. Most of the independent variables from economic, social, and medical field are overlapping with each other, thus, estimating the parameters of the Tobit model may lead to are biased and inconsistent estimation. In order to overcome this issue, a set of methods can be deployed, such as principal component. In this paper, the Bayesian technique will be used for estimating the parameters of the model. The case study included in the paper is the medical problem of the abortion within Iraqi women. The estimation was performed by using Bayesian Tobit principal component regression model through building algorithm in programing language (R).

Full Text: PDF     DOI: 10.15640/arms.v4n2a7