MULTIPLE REGRESSION ANALYSIS OF LANDED PROPERTY VALUES

Abstract

The objective of this study is to identify various factors that influence the transaction price of landed residential properties, namely detached and terrace houses. The rO factors used were classified in three subgroups of physical "i . . factors, surrounding ^amenities and accessibility to various means of public transport. This study was done through developing two landed property value models, one for each category of residential property. Multiple regression analysis was used to select the significant factors to form the regression eguation. Factors selected by multiple regression analysis were tested for multicollinearity. The testing of significance of individual factors was also done. The testing of predictability of the regression eguations serves to confirm the usefulness of the eguation to predict the transaction price based on the given factors. The study concludes the significance of physical factors in influencing the transaction price. The selection of post office as a new factor by regression analysis is also of interest to future studies.