Abstract
This paper tests market efficiency and return predictability for twelve emerging and four developed securitized real estate markets from 1992 through 2009. The analysis is based on autocorrelation tests, as well as both single variance and multiple variance ratio tests. Furthermore, non-parametric runs tests are conducted. Empirical evidence shows that the efficient market hypothesis in its weak form is not rejected by any statistical test for seven of the twelve markets. This result is surprising, since all four developed securitized real estate stock markets analyzed do not follow a random walk. The results are confirmed by the analysis of excess returns following from technical trading rules.