Estimating Internet Users' Demand Characteristics

Abstract

We investigate two parametric approaches and one non-parametric approach toestimating Internet users' value-of-time, an important characteristic ofdemand for Internet services. The advantages of these approaches are madeclear and their limitations discussed. The models are tested with datagenerated from our similation model of the Internet economy. Given thecharacteristics of the data, we investigate parametric count-data modelsfirst. While reasonably good results are obtained on all medium- andlarge-sized jobs, the method fails on small-sized ones. Second, we develop anonparametric quasi-Bayesian update algorithm for retrieving the underlyingdistribution function of Internet users' value-of-time purely fromobservations on their choices. Compared with the parametric count-data models,good results are obtained in every case.