以品牌實際選購行為建立品牌競爭圖之研究

Constructing Competitive Map by Using Mixture Cluster Multinomial Logistic Regression Model

任立中(Li-Chung Jen) ； 陳靜怡(Ching-I Chen) ； 陳成業(Chen-Yueh Chen)

《台灣管理學刊》 5卷2期 (2005/08) Pp. 169-185

https://doi.org/10.6295/TAMJ.2005.0502.01

摘要

顧客表現在品牌選擇行為上的異質性與動態性可確切的反映出品牌之間的競爭態勢。長久以來，競爭態勢的分析經常侷限於定性的問卷調查，得到的只是顧客的主觀認知，而非實際的購買行為。本文擬以廠商所擁有的客戶交易資料庫為研究對象，根據顧客的實際品牌購買行為探究不同品牌之競爭激烈程度。有別於問卷調查所使用的多變量統計模式僅能衡量跨顧客的變數關係，資料庫行銷所使用的計量模式必須得以分析個別顧客本身的動態交易資料，方能衡量購買行為之異質性。本文所採用的混合分群多項式邏吉斯迴歸模式，可衡量顧客表現在品牌選擇行為上的異質性，並進而異中求同產生市場區隔，有助於行銷策略之研擬。顧客的品牌選擇行為能夠反映品牌間之替代與競爭程度，其所形成的品牌競爭圖亦能使廠商有效的辨認主要競爭對手。最後，本文透過表面無相關迴歸模式描述市場區隔的人口統計特質，有助於廠商瞭解新進顧客之所屬市場區隔，使行銷策略之發展更趨完整。

關鍵字

競爭圖 ； 隨機效用模式 ； 混合分群多項式邏吉斯迴歸模式

並列摘要

The heterogeneity and dynamic in brand choice behavior exactly imply the competitive state between brands. For a long time, the measurement of competitive state has been limited to qualitative questionnaire survey, from which we can only get the information of subject recognition but not actual purchase behavior of customers. Therefore, this paper analyzes the actual brand choice behavior recorded in customer database in order to capture the extent of competition among brands.
The econometrics models used in database marketing are distinct from multivariate statistic models used in questionnaire survey. Models for questionnaire survey only can detect the relationship between variables across customers. However, models for database marketing have to be able to analyze each customer own purchase records to capture heterogeneity. We use mixture cluster multinomial logistic regression model to measure the customers' heterogeneity in brand choice behavior and segment the customers into four clusters. Competitive map derived from brand choice behavior can illustrate the degree of substitution and competition among brands and help firms identify their main opponents more efficiently. At last, we use seemly unrelated regression (SUR) model to detect the demographic characteristics of each segment to facilitate marketing persons to segment new customers.

並列關鍵字

Competitive Map ； Random Utility Model ； Mixture Cluster Multinomial Logistic Regression Model

參考文獻

Allenby, G. M(1989).A Unified Approach To Identify Estimating And Testing Demand Structures With Aggregate Scanner Data.Marketing Science.8(3),265-280.

Allenby, G. M,N. Arora,J. L. Ginter(1995).Incorporating Prior Knowledge into the Analysis of Conjoint Studies.Journal of Marketing Research.32(May),152-162.

Allenby, G. M,N. Arora,J. L. Ginter(1995).Using Extremes to Design Products and Segment Markets.Journal of Marketing Research.32(November),392-403.

Andrews, R. L.,A. Ansari,I. S. Currim(2002).Hierarchical Bayes versus Finite Mixture Conjoint Analysis Models: A Comparison of Fit, Prediction, and Partworth Recovery.Journal of Marketing Research.39(February),87-98.

Bucklin, R. E.,S. Gupta(1992).Brand Choice, Purchase Incidence, and Segmentation: An Integrated Modeling Approach.Journal of Marketing Research.29(May),201-215.