Application of a Markov model to study the marketing variables and their individual effects on market share

Date of Publication


Document Type


Degree Name

Doctor in Business Management


Ramon V. Del Rosario College of Business


Business Management

Thesis Adviser

Rao, Purba

Defense Panel Chair

Concepcion, Antonio

Defense Panel Member

Javier, Wilfrido R.
Perez, Errol


Markov models, developed and applied for many years, can be useful to predict market shares and determine marketing strategies. Some of these models try to capture the influence of certain marketing mix variables (product, price, place and promotion) on the consumers' choice behavior. Kotler (32) suggested another type of Markov model which he called The Competitive Marketing Mix Model . This model can be used to predict the effect on the market shares through the improvement of certain marketing variables. The rationale is that if marketing managers have such a model available, they can theoretically determine which marketing variable is the most financially efficient to improve. The model can also provide marketing managers with another tool to help them formulate appropriate marketing strategies. This study attempts to apply the idea described above in the Philippines. Since Markov based models only deal with the frequently bought, inexpensive, everyday products, coconut-based cooking oil was chosen to build and test the model. Only three leading brands (Minola, Golden Fry and Baguio) were included in the study. All other brands and undbranded were categorized as Others . Respondents were 200 housewives in Metro Manila, divided into their respective classes (AB, C and DE). The survey questionnaire was designed to capture perceptions of variables/attributes of product, capture brand switching patterns, and provide the market shares of the selected brands.
Study result shows that a certain degree of brand switching did take place among the cooking oil consumers during the 6-month study period. It can be concluded that switching mainly took place because of non-availability of favorite brand. Another reason for brand switching is price. Consumers, in general, give great attention to the quality of the oil in terms of color and smell. However, due to financial situations the poorer tend to buy high quality cooking oil only sporadically. Advertising is not considered to be important. The main finding of this study concerns the structure of the scale used to rate the variables/attributes and the selected brands. However, result show that the scale design could not capture extreme differences in preferences. Therefore, the model could not be validated. The study concluded that the model can not be established to study marketing variables and their influence on market share. The study noted that the significance of the scale was not discovered during the pre-testing of the questionnaire and was only realized when data was being analyzed. However, another attempt with a modified questionnaire and an improved scale was made. It is recommended that a panel of respondents should be established and that the questionnaire and scale design be further studied.

Abstract Format




Accession Number


Shelf Location

Archives, The Learning Commons, 12F Henry Sy Sr. Hall

Physical Description

207 leaves ; 28 cm.


Market share.; Markov processes.

This document is currently not available here.