Date of Award

Spring 2004

Document Type


Degree Name

Doctor of Philosophy (PhD)


Business Administration-Finance

Committee Director

Kenneth Yung

Committee Member

Mohammad Najand

Committee Member

Vinod Agarwal


This dissertation includes three essays that study on Real Estate Investment Trusts (REITs). The first essay examines the shape of market demand curve by employing REIT equity fund flows as a proxy for REIT aggregate demand. An empirical framework is introduced to disentangle the price-pressure effect and the information effect. We do not find evidence for the price-pressure effect, which states that REIT equity fund flows directly affect REIT market prices. Instead, we find that the performance of REIT market prices does influence REIT equity fund flows. The results indicate that investors adjust their demand for REITs by investigating and forecasting REIT fundamentals. Thus our findings support the horizontal market demand curve for REITs.

The second essay examines the time varying risk premium for equity REITs with both GARCH and GARCH-M specifications. Using a daily data set from June 1995 to September 2002, we find that both the ARCH and GARCH effects are significant for our estimations. The results show that the market returns and the first order autocorrelation help explain the excess returns of equity REITs. However, the movement of interest rates contributes to equity REIT returns only when the market returns are not considered at the same time. Finally, the GARCH-M terms are not significant in determining the expected returns.

The third essay investigates the effects of investor sentiment and institutional ownership on REIT risk premium. Our results show that REIT risk premium is negatively related to investor sentiment. That is, REIT risk premium increases (decreases) when investors become irrationally optimistic (pessimistic). However, we do not find any significant relation between REIT risk premium and institutional ownership changes except for small-size REITs. The results are robust when we include other control variables in the analysis. We also find evidence that the information contained in past investor sentiment changes has an impact on subsequent REIT risk premium. The information contained in past institutional ownership changes, however, only affects REIT risk premium in the time period between 1994 and 1998, and for the good performing REITs.