Traditional finance and behavioral finance: Competition or co-existence? The case of closed -end funds

The main objective of this dissertation was to contribute to the long-standing debate between the traditional and behavioral finance paradigms by examining closed-end funds discounts and returns with multifactor models that included factors from both paradigms. Using the generalized method of moments estimation, this research study provided a more comprehensive examination of the potential traditional and behavioral finance determinants of closed-end funds discounts and returns for the major CEF segments. From traditional finance, agency costs, tax liability, and dividend yield hypotheses were considered for CEF discounts. From behavioral finance, the investor sentiment hypothesis was considered for both discounts and returns.;Based on the results obtained, the investor sentiment proxy was highly statistically significant, with the expected sign, for all samples and segments, and in both simple regression and multifactor model. Also, in the closed-end funds return analysis, the sentiment risk factor from behavioral finance enters the return-generating process of CEF consistently and with the expected sign for all CEF segments, with the exception of country funds, in which it did not enter with the expected sign. The only traditional finance variable that seemed to have some explanatory power for CEF discounts was expense ratio. The main finding of this dissertation, that investor sentiment plays a prominent role in explaining both CEF discounts and returns, constitutes an important result for proponents of behavioral finance by providing support to the investor sentiment hypothesis as a possible explanation of the closed-end fund puzzle and as one of the results that contradicts the traditional approach to finance.