Studying household purchasing and nonpurchasing behaviour for a frequently consumed commodity: two models†

Abstract

A panel data double-hurdle model is developed to analyse household purchasing behaviour. The model, a time-series extension of Cragg's conventional double-hurdle model of censored consumption, is able to account not only for the censored nature of commodity purchases, but also for the temporal linkage of the purchase process. The panel data double-hurdle model is compared with the marked renewal model that has also been used for studying household purchasing behaviour. The empirical results of the double-hurdle model show that for household milk purchases hurdle due to noneconomic reasons exists. The results of the marked purchase renewal model showed that the duration dependence is positive. Both models fit the data well. † The views presented in this paper are those of the authors and not necessarily those of USDA-ERS.