Article ID: | iaor2000742 |
Country: | United Kingdom |
Volume: | 17 |
Issue: | 3/4 |
Start Page Number: | 175 |
End Page Number: | 208 |
Publication Date: | Jun 1998 |
Journal: | International Journal of Forecasting |
Authors: | Ben-Akiva Moshe, Gershenfeld Shari |
Keywords: | time series & forecasting methods |
Many consumer choice situations involve the selection of optional items or features from a menu. Often, in these situations, predesigned bundles or packages of items are also offered. We present an analytical approach to these menu situations which takes into account the large number of alternatives they represent, the interrelated nature of demand for the individual items, and the impact of bundling of items. The emphasis of our data-collection approach is on providing a realistic choice setting which represents all possible alternatives. In our modelling approach, we employ discrete choice modelling techniques which explicitly capture the full range of these alternatives. We present the results of an empirical study of enhanced custom-calling features for residential telephone service. Using this approach, we developed well-behaved choice models that were used to predict demand and revenues for the features under a wide range of marketing scenarios.