PANEL PROBIT WITH FLEXIBLE CORRELATED EFFECTS: QUANTIFYING TECHNOLOGY SPILLOVERS IN THE PRESENCE OF LATENT HETEROGENEITY

SUMMARY In this paper, we introduce a Bayesian panel probit model with two flexible latent effects: first, unobserved individual heterogeneity that is allowed to vary in the population according to a nonparametric distribution; and second, a latent serially correlated common error component. In doing so, we extend the approach developed in Albert and Chib (Journal of the American Statistical Association 1993; 88: 669–679; in Bayesian Biostatistics, Berry DA, Stangl DK (eds), Marcel Dekker: New York, 1996), and in Chib and Carlin (Statistics and Computing 1999; 9: 17–26) by releasing restrictive parametric assumptions on the latent individual effect and eliminating potential spurious state dependence with latent time effects. The model is found to outperform more traditional approaches in an extensive series of Monte Carlo simulations. We then apply the model to the estimation of a patent equation using firm-level data on research and development (R&D). We find a strong effect of technology spillovers on R&D but little evidence of product market spillovers, consistent with economic theory. The distribution of latent firm effects is found to have a multimodal structure featuring within-industry firm clustering. Copyright © 2012 John Wiley & Sons, Ltd.

[1]  S. Chib,et al.  Bayesian analysis of binary and polychotomous response data , 1993 .

[2]  T. Lancaster.  An Introduction to Modern Bayesian Econometrics , 2004 .

[3]  John M. Olin.  On MCMC sampling in hierarchical longitudinal models , 1999 .

[4]  Jason Abrevaya.  Leapfrog estimation of a fixed-effects model with unknown transformation of the dependent variable , 1999 .

[5]  Keisuke Hirano,et al.  Semiparametric Bayesian Inference in Autoregressive Panel Data Models , 2002 .

[6]  The mean and right-tail probability estimation in a Pareto distribution , 2003 .

[7]  M. Harding,et al.  A Bayesian mixed logit-probit model for multinomial choice , 2008 .

[8]  Steven Berry,et al.  Nonparametric Identi cation of Multinomial Choice Demand Models with Heterogeneous Consumers , 2008 .

[9]  J. Geweke,et al.  Contemporary Bayesian Econometrics and Statistics , 2005 .

[10]  Markku Rahiala,et al.  Panel Data , 2011, International Encyclopedia of Statistical Science.

[11]  Stéphane Bonhomme,et al.  Robust priors in nonlinear panel data models , 2007 .

[12]  Fabrice Defever.  Quantitative economics 2 , 2010 .

[13]  Siddhartha Chib,et al.  Bayes regression with autoregressive errors : A Gibbs sampling approach , 1993 .

[14]  Gary Chamberlain,et al.  Chapter 22 Panel data , 1984 .

[15]  Jeffrey M. Woodbridge.  Econometric Analysis of Cross Section and Panel Data , 2002 .

[16]  Simon J. Godsill,et al.  Bayesian model selection for time series using Markov chain Monte Carlo , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[17]  Radford M. Neal.  Markov Chain Sampling Methods for Dirichlet Process Mixture Models , 2000 .

[18]  Siddhartha Chib,et al.  Semiparametric Bayes analysis of longitudinal data treatment models , 2002 .

[19]  R. Paap.  What are the advantages of MCMC based inference in latent variable models? , 2002 .

[20]  G. Chamberlain.  Multivariate regression models for panel data , 1982 .

[21]  Greg M. Allenby,et al.  Modeling Household Purchase Behavior with Logistic Normal Regression , 1994 .

[22]  Denis Bolduc,et al.  Multinomial Probit Estimation of Spatially Interdependent Choices: An Empirical Comparison of Two New Techniques , 1997 .

[23]  David B. Dahl,et al.  Sequentially-Allocated Merge-Split Sampler for Conjugate and Nonconjugate Dirichlet Process Mixture Models , 2005 .

[24]  Jinyong Hahn,et al.  A likelihood-Based Approximate Solution to the Incidental Parameter Problem in Dynamic Nonlinear Models with Multiple Effects , 2016 .

[25]  K. Train.  A Comparison of Hierarchical Bayes and Maximum Simulated Likelihood for Mixed Logit , 2001 .

[26]  Jerry A. Hausman,et al.  Panel Data and Unobservable Individual Effects , 1981 .

[27]  Christopher R. Knittel,et al.  Estimation of Random Coefficient Demand Models: Challenges, Difficulties and Warnings , 2008 .

[28]  M. Escobar,et al.  Bayesian Density Estimation and Inference Using Mixtures , 1995 .

[29]  John Geweke,et al.  Efficient Simulation from the Multivariate Normal and Student-t Distributions Subject to Linear Constraints and the Evaluation of Constraint Probabilities , 1991 .

[30]  Mark A. Schankerman,et al.  Identifying Technology Spillovers and Product Market Rivalry , 2005 .

[31]  Siddhartha Chib,et al.  Inference in Semiparametric Dynamic Models for Binary Longitudinal Data , 2006 .

[32]  Ivan Fernandez-Val,et al.  Fixed effects estimation of structural parameters and marginal effects in panel probit models , 2007 .

[33]  John Geweke,et al.  Contemporary Bayesian Econometrics and Statistics: Geweke/Contemporary Bayesian Econometrics and Statistics , 2005 .

[34]  A. Jaffe.  Technological Opportunity and Spillovers of R&D: Evidence from Firms&Apos; Patents, Profits and Market Value , 1986 .

[35]  Kenneth E. Train,et al.  Discrete Choice Methods with Simulation , 2016 .

[36]  J. Richard,et al.  Classical and Bayesian Analysis of a Probit Panel Data Model with Unobserved Individual Heterogeneity and Autocorrelated Errors , 2008 .

[37]  Sokbae Lee,et al.  Is Distance Dying at Last? Falling Home Bias in Fixed Effects Models of Patent Citations , 2007 .

[38]  Tong Li,et al.  SEMIPARAMETRIC BAYESIAN INFERENCE FOR DYNAMIC TOBIT PANEL DATA MODELS WITH UNOBSERVED HETEROGENEITY , 2008 .

[39]  R. Kohn,et al.  Bayesian Estimation of a Random Effects Heteroscedastic Probit Model , 2008 .