A Semi-Nonparametric Estimator for Random Coefficient Demand Models
2017/03/02
研討會日期 : 2017-03-02
時間 : 14:30
主講人 : Professor Jing Tao
主持人 : Professor Chu-An Liu
地點 : Conference Room B110
演講者簡介 : Professor Tao received her Ph.D. in Economics from University of Wisconsin-Madison in 2015. She is currently an Assistant Professor at University of Washington. Her primary research field is Econometrics, and secondary research fields are Empirical Industrial Organization, Applied Microeconomics.
演講摘要 : In this paper, we develop a simple two-step nonparametric estimator for the widely used random coefficients logit demand models. The first step obtains choice-specific errors as the residuals from a simple nonparametric IV regression of (logarithm of) quantity on product characteristics; the second step treats the errors as data and employs MLE to estimate the distribution of random coefficients nonparametrically. Comparing to the standard nested fixed point GMM estimator, ours relaxes the parametric assumptions on the distribution of random coefficients and is computationally attractive as it avoids solving fixed points of demand systems. We establish the large sample properties of the estimator and Monte Carlo simulations confirm the theoretical results. We also consider extensions to dynamic discrete choice models. Finally, we apply the proposed method to nonparametrically estimate the distribution of consumers' willingness-to-pay for organic/health products using supermarket scanner data.
