A multiple linear regression analysis in determining the significant factors explaining bank profitability

• Main Authors: Ang, Cristine R., Vasquez, Estela A.
• Format: text
• Language: English
• Published: Animo Repository 1997
• Subjects: Regression analysis > Data processing; Analysis of covariance; Log-linear models; Banks and banking
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/16431
• Institution: De La Salle University

Multiple linear regression analysis is a statistical technique that can be used to analyze the relationship between a single dependent variable and several independent variables (Hair, Anderson, Tatham, and Black, p. 24). This thesis is an application of this statistical tool in the field of banking. The researchers' main objective was to determine the factors significantly explaining the profitability of four types of banks, namely commercial, development and savings banks, as well as savings and loan associations. After the data collection, the forward stepwise procedure, through the Statistical Analysis System (SAS) was utilized to develop the subset of independent variables to be included in the regression model. The researchers obtained four tentative models, one for each type of bank with the annual profit as the dependent variable. For commercial banks, there were five factors included. For development banks, there were two, namely amount of loan portfolio and amount of cash. Savings banks included the same factors as that of the development banks, together with total capital accounts. Lastly, savings and loan associations included three factors, namely cash, deposits, and accounts. Analysis of the obtained tentative models was then performed by the researchers to verify assumptions made regarding the components of the model. Eventually, the researchers ended up with the tentative models as the final regression models. Finally, these models were further analyzed and interpreted in terms of underlying banking concepts and principles.