Abstract
This paper proposes a pairwise likelihood estimator based on an analytic approximation method for the random effects probit model. It is widely known that the standard approach for the random effects probit model relies on numerical integration and that its likelihood function does not have a closed form. When the number of time periods or the serial correlation across periods is large, the resulting estimator is likely to become biased. This study derives an analytic approximation for the likelihood function of one pair of time periods without relying on typical numerical-integral procedures. We then apply this formula in a pairwise likelihood estimation procedure to derive our estimator, which is obtained as the product of the analytic approximation of the likelihood function for all possible pairs of time periods. A simulation study is conducted for the comparison of our proposed estimator with the estimators for the pooled probit model and Gaussian quadrature procedure. The evidence shows that our proposed estimator enjoys desirable asymptotic properties. In addition, compared to the estimator based on the Gaussian quadrature procedure, our proposed estimator exhibits comparable performances in all the configurations considered in the simulation study and shows superiority for the cases of a large number of time periods and high serial correlation across periods. We apply our proposed estimator to British Household Panel Survey data so as to characterize the trend of working probabilities.
Similar content being viewed by others
Notes
Specifically, Borjas and Sueyoshi (1994) and Lee (2000) study the random effects probit model with group effects. While the standard random effects panel model concerns serial correlation at the individual level, the model with group effects focuses on the correlation between individuals within a group. In addition to “group”, the terminology of “cluster” or “unit” is also commonly used. Therefore, the number of time periods and the number of individuals in the panel data model are parallel to the number of individuals and the cluster size in the group effects model, respectively.
The detailed proof can also be found via online supplementary files and the website, http://idv.sinica.edu.tw/wjtsay/htm/jen02a.htm.
In the cases of \(N=100,\rho =0.95\) as well as \(N=1000,\rho =0.50\), the successful convergence rates are 97.0476 and 97.4286%, respectively.
The only one exception is the case of \(T=5, N=2000\), and \(\rho =0.5\) where the successful rate of normal convergence is 99.90%. Table 4 summarizes the successful rates of normal convergence for all three estimators.
The results for the female sample are not reported because of space limitation, which can be found via the website, http://idv.sinica.edu.tw/wjtsay/htm/jen02a.htm and online supplementary files.
References
Abramowitz M, Stegun IA (1972) Handbook of mathematical functions with formulas, graphs, and mathematical tables, 10th edn. Dover, New York
Aigner D, Lovell CAK, Schmidt P (1977) Formulation and estimation of stochastic frontier production function models. J Econom 6:21–37
Arnold BC, Strauss D (1991) Pseudolikelihood estimation: some examples. Sankhy\(\overline{a}\) Indian J Stat Ser B 53:233–243
Besag JE (1975) Statistical analysis of non-lattice data. J R Stat Soc Ser D 24:179–195
Booth AL, Jenkins SP, Serrano CG (1999) New men and new women? A comparison of paid work propensities from a panel data perspective. Oxf Bull Econ Stat 61:167–197
Borjas GJ, Sueyoshi GT (1994) A two-stage estimator for probit models with structural group effects. J Econom 64:165–182
Butler JS, Moffitt R (1982) A computationally efficient quadrature procedure for the one-factor multinomial probit model. Econometrica 50:761–764
Cox DR, Reid N (2004) A note on pseudolikelihood constructed from marginal densities. Biometrika 91:729–737
Guilkey DK, Murphy JL (1993) Estimation and testing in the random effects probit model. J Econom 59:301–317
Heagerty PJ, Lele SR (1998) A composite likelihood approach to binary spatial data. J Am Stat Assoc 93:1099–1111
Heckman JJ, Willis RJ (1976) Estimation of a stochastic model of reproduction: an econometric approach. In: Terleckyj NE (ed) Household production and consumption. National Bureau of Economic Research, New York
Kuk AYC, Nott DJ (2000) A pairwise likelihood approach to analyzing correlated binary data. Stat Probab Lett 47:329–335
Lee LF (2000) A numerically stable quadrature procedure for the one-factor random-component discrete choice model. J Econom 95:117–129
Mardia KV, Kent JT, Hughes G, Taylor CC (2009) Maximum likelihood estimation using composite likelihoods for closed exponential families. Biometrika 96:975–982
Press WH, Teukolsky SA, Vetterling WT, Flannery BP (1992) Numerical recipes, 2nd edn. Cambridge University Press, New York
Renard D, Molenberghs G, Geys H (2004) A pairwise likelihood approach to estimation in multilevel probit models. Comput Stat Data Anal 44:649–667
Stroud A, Secrest D (1966) Gaussian quadrature formulas. Prentice-Hall, Englewood Cliffs
Tsay WJ, Huang CJ, Fu TT, Ho IL (2013) A simple closed-form approximation for the cumulative distribution function of the composite error of stochastic frontier models. J Product Anal 39:259–269
Varin C, Vidoni P (2009) Pairwise likelihood inference for general state space models. Econom Rev 28:170–185
Author information
Authors and Affiliations
Corresponding author
Additional information
The authors thank Stephen P. Jenkins for providing data used in Booth et al. (1999).
Electronic supplementary material
Below is the link to the electronic supplementary material.
Appendix
Appendix
Rights and permissions
About this article
Cite this article
Lo, TF., Ke, PH. & Tsay, WJ. Pairwise likelihood inference for the random effects probit model. Comput Stat 33, 837–861 (2018). https://doi.org/10.1007/s00180-017-0757-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00180-017-0757-8