- 學位論文
Applying Bayesian Statistical Model in Competitive Market Structure under Incomplete Information：an Empirical Study of IT Industry
Applying Bayesian Statistical Model in Competitive Market Structure under Incomplete Information：an Empirical Study of IT Industry
並列摘要
On the intensely competitive market landscape, a complete market structure information is urgently indispensable for marketing managers to set critical marketing strategies. In other words, market structure provides complete market information, and is essential for formulation of integrated marketing strategies. Most of marketing practitioners, however, encounter the unavailability of complete market structure data. In order to overcome this obstacle, market research companies are emerging (such as A.C. Nielson, Gartner etc.) in recent years. Since complete market structure data require lots of time to investigate, collect, and analyze, thus the analyst of research companies usually faces the dilemma of accuracy or efficiency. Therefore, how to accurately measure market structure is a topic analysts must investigate in depth. In order to improve the accuracy and efficiency of the estimation method that market research companies usually use, this thesis provides a systematic methodology for estimating market structure data. This thesis develops two methods to revise the second-period estimated relation matrix values. One is naïve method of direct substitution and the other is Bayesian statistical analysis. By using these two methods presented by the thesis, there is considerable evidence to prove that Bayesian statistical Model can effectively improve the accuracy and efficiency of estimation. Finally, we hope that this method can help the marketing manager develop appropriate marketing channel and segmentation strategies.