An Analysis of Preference Heterogeneity in Residential Choice Behaviour

Abstract

In the present, the public transportation via rail are significantly developed and covering in many areas. Currently, railway development has already been covered over 153 kilometers. This has given the influence on land use, creating more real estate development along the route. It is necessary to know amount of travel demand and populations in area precisely. To know this number, it is required a comprehension of various population behavior in choosing residence as an important factor. Hence, the objective of this study is to learn the various behaviors of each group of people with different habitat characteristics or the heterogeneity preference. Since each people has different taste, preference, interest, and lifestyles of choosing their own residences. The researcher considered that ‘Homogeneous selection’ analysis through the method of ‘Discrete choice model’ was not giving an accurate result to the behavior of choosing residence in the real situation. In this research, it is using 3 compared models including, Multinomial Logit model (MNL), Mixed Logit model (MMNL), and Latent class model (LC). Researcher checked Goodness of fit of mentioned model via Maximum likelihood method. The data of this analysis gather a sample survey. The total sample was 2,400 samples from people who lived in Bangkok and Nonthaburi province by creating a scenario (Stated preference or SP) of housing selection. However, the analysis illustrated that each persons’ housing selection behavior are statistically different. We divided people with similar characteristic into a group of economic, and social factor. Example of sub divided group included, number of members in household, average income, and status of each respondent.