Deveolpment of a demand forecasting model for new product and its applications to telecommunications market

Abstract

Telecommunications service market is changing markedly. Many once monopolistic markets are becoming more competitive and the tendency is for these transformations to occur more quickly and frequently. To demonstrate and forecast the complicated demand patterns in a quickly changing telecommunications market, models must address the interrelationships among services and/or products. Much effort in recent years has been focused on developing aggregate diffusion models incorporating these interrelationships. One of the purposes of this study is to apply such a diffusion model to the telecommunications market. Consumers choose the particular service that maximizes their utility and efficiently uses their limited resources; they adopt new products whenever they can maximize their utility. Thus sales patterns of new products result from choices made by consumers over time. Therefore, by studying the choice behavior of consumers, we can understand the sales patterns of new products and develop sales forecasting models for those products. This study develops new product diffusion models in several different environments based on consumer choice behavior. Then we incorporate diffusion effects and choice effects in an integrated model. The choice literature generally ignores demand dynamics and previous diffusion models rarely include control variables. The proposed model is a combination of the two approaches. The basic premise of the proposed models is that sales patterns of new products are based on the choice behavior of consumers, where consumers choose a product to maximize their utility. Then we can derive the implied relationships among choice probabilities, diffusion processes, and marketing mix variables. The proposed model differs from previous diffusion models in that effects of exogenous variables such as price are incorporated into the diffusion processes through the choice behavior of the consumer. First, we develop a choice-based competitive diffusion ...