Finance - Theses
Permanent URI for this collection
Search Results
1 - 3 of 3
-
ItemThe holiday effect in Asian stock markets: spending holidays, non-spending holidays and firm size( 2013)Calendar-based seasonalities in stock markets have attracted considerable interest among academics as well as practitioners and have been the subject of much empirical research. These seasonalities, or anomalies, refer to the proposition that stock returns exhibit regular and predictable changes that tend to recur over long periods of time. This thesis examines the effects of two calendar seasonalities, namely the holiday effect and the January effect, over a 22-year period in the stock markets of Hong Kong, Indonesia, Malaysia, Singapore and Taiwan. It provides strong evidence that only those holidays associated with heavy consumer spending affect return patterns. By exploiting the fact that the timing of many festivals in Asia varies relative to the Gregorian calendar, the thesis is also able to separate year-end effects from holiday effects. It is also found that the effects are stronger in smaller firms. This thesis is the first to test simultaneously for calendar effects in five Asian markets using daily stock prices collected from one data source. Furthermore, it is the first to test calendar effects at the firm level (rather than at the market index level) in four of the five countries.
-
ItemEssays on institutional investors' behaviour and performance( 2013)Chapter 1: On Market States and the Value of the Actively Managed Mutual Fund Industry This paper presents a novel approach to quantify the economic value of mutual funds’ conditional performance in bear market states. In particularly, this approach jointly evaluates fund managers’ market timing and selectivity skills across market states. After accounting for the insurance premium of managers’ conditional performance in bear states, this study finds that the extra benefit arising from active mutual fund management in bad times could largely compensate for its cost at the aggregate level. Chapter 2: Mutual Fund Performance, Momentum Trading and January Seasonality This paper demonstrates that the actively managed U.S. equity mutual fund does not underperform the CAPM, Fama–French three-factor, and Carhart four-factor benchmarks during 11 months of the year. In contrast, mutual funds considerably underperform the passive benchmarks in January. We find that the January underperformance is due largely to mutual fund managers chasing momentum trends, which experiences a dramatic reversal in January. Since fund managers disproportionally overweight medium-cap winners at the end of the year, the subsequent reversal of these past winners significantly drags down the average returns of mutual funds in January.
-
ItemEssays on momentum investing( 2013)Chapter 1: This paper reexamines the apparent success of two prominent stock trading strategies: long-term contrarian and intermediate-term momentum. The paper demonstrates that long-term contrarian is entirely attributable to the classic January size effect, rather than to investor overreaction, as argued by De Bondt and Thaler (1985). Further, the paper also resolves the Novy-Marx (2012) concern about whether return autocorrelation “is really momentum” by demonstrating that the superior performance of intermediate-term momentum is due to strong January seasonality in the cross-section of returns. The implications are that long-term contrarian must be considered largely illusory, and intermediate-term momentum must take account of annual seasonalities in returns. Chapter 2: Macroeconomic risk continues to be proposed as a source of stock price momentum. For instance, Liu and Zhang (2008) claim that the growth rate of industrial production “plays an important role in driving momentum profits”. This paper shows that the growth rate of industrial production is not the source of momentum profits. Because recent winners and recent losers have nearly identical loadings on the growth rate of industrial production outside of January, there is essentially a net zero factor loading in the 11 months of a year when momentum does exist, and a difference only in January, when losers massively outperform winners. We also document the fact that the growth rate of industrial production is not a priced risk factor outside of January. Moreover, application of the same methodology to all factors reveals no evidence that an explanation for momentum profits lies in macroeconomic risk.