Abstract
Changes in relative price and dealing activity are likely to affect different segments of the market in different ways. A distributed lag model is developed for predicing a firm's market share over a period of weeks from knowledge of the changes in these variables. It is tested on aggregate data for a metropolitan market, and applied to data for three classes of underlying segments. The bases for segmenting the market are by family purchasing characteristic, package size, and channel of distribution.
Get full access to this article
View all access options for this article.
