RETURN PREDICTABILITY: DOES IT EXIST IN SINGAPORE REITS?

Abstract

This paper examines the characteristics of REITs and its performance in Singapore. Its main aim is to investigate the return behaviour on S-REITs, discerning whether their future returns can be predicted. Mirroring the research models used in Welch and Goyal (2008) and Rapach et al. (2009), the paper has incorporated the use of predictive regression models and remodels them into the S-REIT market’s context. The in-sample observations lay the groundwork, which the future return predictions are based on. Multiple out-of-sample periods are proposed to check the robustness of the estimated future returns. The forecasted results for the out-of-sample periods are evaluated using specialized metrics, testing the existence of return predictability in S-REITs and its statistical significance. The paper has produced sound evidence on the existence of return predictability in Singapore. The research findings also prove that foreign concepts and research methods can be applied to Singapore’s market conditions. Recognised valuation ratios of the stock market index and economic indicators has fielded consistent results in Singapore when compared to their overseas performance. With better statistical scores, combinations techniques continue to outshine the individual predictive regression models. The Kitchen Sink model acts as a comparison for the combined schemes. Its underperformance in this paper is no outlier as previous literature documented likewise. Although the paper has a successful research outcome, it is important to note that the research methodology in this paper has its own constraints. Further research is proposed when such limitations are no longer in place.