MARKET SHARE MODELLING AND FORECASTING USING MARKOV CHAINS AND ALTERNATIVE MODELS

摘要

Markov chain has been a popular approach for market share modelling and forecasting in many industries. This paper presents four mathematical models for the same market share problem based on different underlying assumptions. The four models include a homogeneous Markov model, a time-varying Markov model, a new extended time-varying Markov model, and a novel non-Markov model. A numerical example in the telecommunications industry is included to illustrate that all four models can be used for market share forecasting. Although Markov models are popular, forecasters should be cautious in choosing Markov or the alternative models for their problems in hand. In order to achieve the best forecasting results, forecasters should have in-depth understanding of the industries, market conditions, and trends, then make appropriate assumptions, and apply or even develop the most suitable models.