TY  - RPRT
        AU  - Jenkins, Stephen P.
AU  - Rios-Avila, Fernando
        TI  - Fitting the Bivariate Mixed Poisson Regression Model by Maximum Simulated Likelihood
        PY  - 2026/Apr/
        PB  - Institute of Labor Economics (IZA)
        CY  - Bonn
        T2  - IZA Discussion Paper
        IS  - 18606
        UR  - https://www.iza.org/publications/dp18606
        AB  - We introduce bimpoisson, a program to fit the bivariate mixed Poisson regression model by maximum simulated likelihood (MSL) as per Munkin and Trivedi (The Econometrics Journal, 1999). Options include: sampling function or standard MSL; pseudo-random uniform or Halton draws, antithetic acceleration, and a first-order bias correction. We also provide post-estimation tools to predict conditional count probabilities and expected counts. We examine bimpoisson’s performance using Monte Carlo simulation analysis and provide two empirical illustrations using data from Xu and Hardin (The Stata Journal, 2016) and Munkin and Trivedi (1999). We provide practical advice about which MSL estimator and types of draws and number to use.
        KW  - st0001
KW  - bimpoisson
KW  - bivariate mixed Poisson regression
KW  - maximum simulated likelihood
KW  - sampling function
KW  - count data
        ER  -