Files
Abstract
We propose an econometric technique for estimating the parameters of a binary choice model when only aggregated data are available on the choices made. The method performs favorably in applications to both simulated and real world choice data.