The Sales Forecasting In High-Technology Markets Based On Utility Approach

Accurate sales projections are of vital importance to the profitability and long- term survival of companies, it is an important component of the decision support system in companies also. As the high-tech products diffusion in the market, there is great uncertainty in volatility of market, technology and competition, and differences in consumer behavior significantly have an impact on product sales. At present, diffusion models are commonly adopted to forecast the sales of products in the marketing, such as the Bass model and its expansion model, it's widely used in the retail services, industrial technology, agriculture, medicine, consumer durables, and so on. This model requires that the data of the research object must include the sales peak to ensure the stability of estimation. The managers'interest in sales forecasts, however, is likely to diminish after the sales peak, that is, when the market matures and the high-tech product becomes a commodity. Furthermore, the traditional diffusion modeling approach does not provide a coherent framework to reasonably take into account the industry-specific buying behavior. Since the models have largely been the result of operationally characterizing the theory of innovation diffusion proposed by the sociologist Rogers, their main focus is on the communication aspect. Within this framework, it is not straightforward to consider sales affecting factors other than word-of-mouth recommendation even though there are further phenomena such as price changes of new products, network effects, and forward-looking behavior and consumer heterogeneity, which are directly related to the development of technology- driven markets. Therefore, from the perspective of consumer behavior this paper establishes a new model of products diffusion for the high-technology markets, considering the main factors affecting its diffusion, obviously the performance of the new model will be more reasonable and scientific than before in sales forecasting.After extensive literature search and inspection work, this dissertation summa- rizes many past new product diffusion theories and models both at domestic and foreign, and makes an intensive study of the model structure and parameter estimation methods. Then the influencing factor of product sale in high-technology markets can be interpreted qualitatively after we combine the crossed study of psychology, marketing and diffusion theory and by considering the characteristics of high-tech products. By examining recent studies focusing on consumer buying behavior in high-technology markets, we investigate several main factors as follows can affect consumer decisions, namely, interpersonal communication, price changes of new products, network effects, consumers forward-looking behavior, and consumers heterogeneity. Against this background, we developed a sales forecasting model in high-technology based on utility approach referring to the diffusion model and introduction of the utility approach. This model is specifically used in high-tech products sales forecast, and its advantage is that we enables reliable predictions, in cases of where only short time series are available for calibration. We select the most important of the last two decades innovations in the field of high technology consumer electronics industry on the CD player, DVD player and digital camera as the research object, using the market sales data for empirical study. The empirical comparisons include alternative specifications of the Bass diffusion model as well as a proportional hazard model and consist of two steps. First, the models are checked whether they are able to represent all the sales dataset. Second, the models are applied to predict future sales in the three markets. It is shown that both versions of the proposed model are at least equivalent to the models used as benchmarks in terms of fit. The resulting forecasts show that the proposed model performs significantly better than its benchmarks. Its parsimony enables reliable predictions, even in cases, where only short time series are available for parameter estimation. The model is able to anticipate decreasing diffusion rates as they occur at the end of the growth stage and, thus, helps to avoid overoptimistic sales forecasts, which may cause severe economic damages.