Discrete choice models and heuristics for global nonlinear optimization
The lecture will consist in two parts. First, recent advances in discrete choice models will be presented and motivated. The estimation of these advanced models involves the maximization of a nonlinear, nonconcave loglikelihood function. The nonconcavity of the function has motivated the development of an efficient heuristic to escape from local maxima. This heuristic will be presented during the second part of the lecture, together with promising numerical results.
Record created on 2008-02-15, modified on 2017-02-16