Supplement to "Improving the Numerical Performance of BLP Static and Dynamic Discrete Choice Random Coefficients Demand Estimation"

This appendix discusses the implementation details for MPEC and NFP applied to the BLP demand estimation problem, the KNITRO outputs for MPEC and NFP, how a researcher would adapt static MPEC to a likelihood-based estimation of random-coefficients-logit demand, varying the quality of the data, and dynamic BLP with one consumer type.

Supplemental Authors: 
Jean-Pierre Dub?, Jeremy T. Fox, and Che-Lin Su