The Mixed Birnbaum Model: Estimation using Collateral Information

This paper introduces the Mixed Birnbaum model with collateral information. A simulation study is conducted to evaluate parameter estimation for this model. More specifically, the gain of incorporating collateral information into the model is investigated. The results show that the standard errors as well as latent class assignment can benefit substantially from incorporating external variables that associate with the latent class variable, especially when the sample size is relatively large.