The Conditional Distribution of Real Estate Returns: Are higher moments time varying?

Previous research has shown that the returns on individual properties and listed property securities are skewed (Lizieri and Ward 2001, Young and Graff 1995 and Liu et al. 1992). This claim is investigated in the context of listed UK property companies and US REITs. In particular, the shape of the conditional distribution of total monthly returns is examined for a group of 20 UK companies and 20 REITS listed continuously since 1970 and 1977, respectively. Also investigated is the claim of Young and Graff that the skewness found in property returns varies over time. Using the model of Hansen (1994) it is found that while a large portion of property security returns in the sample do exhibit skewness in the conditional distribution only in a few instances is there evidence of time variation in the skewness parameter. When time varying skewness is found there is little evidence to suggest it is associated with the economic cycle. The link between time varying skewness models and downside risk measures is also discussed and estimates of conditional downside risk are calculated for those companies exhibiting the time varying skewness