Cox Proportional Hazards Regression for Interval Censored data with an Application to College Entrance and Parental Job Loss

Abstract This paper conducted a survival analysis by fitting a Cox proportional hazards model to the Korea Labor panel data in order to analyze the effect of parental job loss on children’s admission to a college and university. Since the Korea Labor panel data is interval-and right-censored, we compared three imputation methods: simple omission, imputation as the average of the left and the right values of the interval, and multiple imputations proposed by Pan (2000). Their integrated area under the curve (iAUC) and mean square error (MSE) were compared in order to assess the predictive and estimation performances. It was found that, within the simulation, the multiple imputation method had a lower MSE than the other two methods. However, no difference was observed in the iAUC values. In the group where each householder had at least a college degree, the parental job loss variable was significantly related to college or university admission of the first child in a household regardless of the interval-censoring imputation method.