Abstract
AbstractHousehold butter expenditures are modeled by market participation and infrequency‐of‐purchase specifications. The market participation model assumes that zero expenditures represent either traditional corner solutions or consumers who never use the product, whereas the infrequency‐of‐purchase model assumes that zero expenditures represent either corner solutions or consumption out of storage. We determine that the infrequency‐of‐purchase specification is preferred to the market participation model by applying a nonnested test based upon likelihood functions. If household expenditure surveys contained information about participation and frequency of purchase, a likelihood function which contained both models could be written and estimated.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have