On the Tradeoff between Computational Simplicity and Asymptotic Properties in Multivariate Probit

被引:6
作者
Kimhi A. [1 ]
机构
[1] Dept. of Agric. Econ. and Management, Faculty of Agriculture, Hebrew University, P.O. Box 12
关键词
Efficiency; Multivariate probit; Quasi maximum likelihood; Simulation;
D O I
10.1023/A:1008649923740
中图分类号
学科分类号
摘要
This paper discusses the most efficient estimator among Quasi Maximum Likelihood Estimators using at most two levels of numerical integration, for the multivariate probit model. Simulations show that this estimator is more efficient but not more costly than the second-best alternative. However, its added efficiency depends on the correlation structure.
引用
收藏
页码:93 / 101
页数:8
相关论文
共 8 条
[1]  
Avery, R.B., Hansen, L.P., Hotz, V.J., Multiperiod probit models and orthogonality condition estimation (1983) International Economic Review, 24, pp. 21-35
[2]  
Avery, R.B., Hotz, V.J., (1985) HotzTran User's Manual, , CERA Economic Consultants, Old Greenwich, CT
[3]  
Kimhi, A., Quasi maximum likelihood estimation of multivariate probit models: Farm couples' labor participation (1994) American Journal of Agricultural Economics, 76, pp. 828-835
[4]  
Lee, L.-F., On efficiency of methods of simulated moments and maximum simulated likelihood estimation of discrete response models (1992) Econometric Theory, 8, pp. 518-552
[5]  
Maddala, G.S., (1983) Limited-Dependent and Qualitative Variables in Econometrics, , Cambridge University Press
[6]  
McFadden, D., A method of simulated moments for estimation of discrete response models without numerical integration (1989) Econometrica, 57, pp. 995-1026
[7]  
Sobel, M.E., Arminger, G., Modeling household fertility decisions: A nonlinear simultaneous probit model (1992) Journal of the American Statistical Association, 87, pp. 38-47
[8]  
White, H., Maximum likelihood estimation of misspecified models (1982) Econometrica, 50, pp. 1-25