Abstract
dc:descriptionThe objective of this thesis is to evaluate the importance of industry portfolios in asset allocation and portfolio choice decisions. Chapters 1 and 2 provide an introduction to the topic and literature.In Chapter 3, we study industry momentum in eighteen developed stock markets from the perspective of a U.S. investor. Momentum strategies are very popular among investment practitioners and some believe that momentum strategies are close to what academics call a “free lunch”. Our findings do not indicate so. Typical industry momentum strategies, once very profitable, appear to be weaker over the last years. Furthermore, investors by following momentum strategies may experience severe and irreversible losses even under favorable conditions in the equity markets. However, our evidence suggests that long-only industry momentum strategies significantly outperform passive strategies both on a raw and a risk-adjusted basis.In Chapter 4, we study the dynamic portfolio choice problem of an investor who faces a time-varying investment opportunity set consisting of industry portfolios that behave differently over good and bad times. We apply a quantitative framework in which investment opportunities and state variables are parameterized using a vector autoregressive model. We find that among industry portfolios, there are significant differences in their performance over changing market conditions as well as in their exposure to different risks over good and bad times a fact suggesting that, besides tactical implications, there may be implications for long-term investors as well. At short horizons, the mean allocation is heavily concentrated in defensive industries which outperform in bad times, but, as the horizon increases, investors allocate significantly more in aggressive industries which outperform in good times. Furthermore, we perform an out-of-sample evaluation of the long-term portfolio model. The results of the performance evaluation suggest that investors who follow dynamic industry rotation strategies can significantly benefit in terms of certainty equivalent returns on wealth and other performance metrics. In addition, the benefits of the proposed dynamic industry rotation strategy are more evident during turbulent times such as the decade of the 2000s.In Chapter 5 we assess the predictive ability of regime-switching models in identifying stock market turning points. In particular, we evaluate an industry rotation strategy in which portfolio adjustments take place after the real-time identification of turning points; the portfolio is invested in aggressive industries during bull markets and in defensive industries during bear markets. We evaluate the economic importance of both homogeneous and heterogeneous models in which transitions are governed by variables identified in the return-predictability literature such as the short rate, the yield spread, the default spread, the dividend yield, and the earnings-to-price ratio.
Degree
thesis:*- Grantor dc:publisher
- University of Piraeus (UNIPI)
- Year dc:date
- 2014
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Malafouris, Vasileios
- Μαλαφούρης, Βασίλειος
Subjects
dc:subject × 12Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/38748