Housing demand in Shanghai: A discrete choice approach

In this study we apply a random utility modeling framework to analyze housing demand in the city of Shanghai. A Multinomial Logit Model taking account of latent choice sets is employed to investigate the impact of household characteristics, such as income, size and age composition, on the choice of dwelling, specified by location, size and unit price. In addition to the price and income effects on housing demand, the model identifies a quality indicator for dwelling attributes, which can be interpreted as the mean attractiveness in a money metric measure.The data used in this study are cross-sectional survey data. The estimated model is used to calculate demand elasticities and demand probabilities, for selected groups of households and types of dwellings. Among the results can be noted that the price-income ratios, the age composition and size of household are all important determinants of the demand. The impact of income distribution on housing demand is also studied.