This thesis involved an empirical investigation of the predictability of Australian industrial stock returns using a dynamic state-space framework. The systematic risks of industrial portfolios were examined in a stochastic market- model. The systematic risks of industry portfolios are found to be stochastic processes. Most of the industry groups have time-varying systematic risks that are mean-reverting to their stable or moving long-term mean. However, the investment and financial services, alcohol and tobacco, gold, insurance and media industry groups have rather random systematic risks. The time-varying market model provides a better explanation of the portfolio returns than the single-index model since it captures the stochastic properties of market risk. Further, a Bayesian dynamic-forecasting model was employed to examine the explanatory power of a set of economic and financial variables. The unanticipated components of the term-structure variable, the interest-rate variable and the aggregate-dividend-yield variable were shown to be significant in explaining the industry portfolio excess returns. The comparison between multivariate analysis and univariate analysis strongly indicates that the correlations within industries are critical in the investigation of the predictability of returns. In the out-of-sample analysis, a maximally predicted portfolio (MPP) was constructed based on the updated economic and financial information; however, the predictability of the MPP did not exceed that of a naive forecast. / Furthermore, the market timing ability associated with the predictability of the MPP was insignificant. The industry-group-rotation strategy is able to enhance the industry portfolio performance, but the predictability only contributes a small proportion of the profits. The results indicate that the industry returns contain predictive components; however, investors are less likely to exploit the existing predictability to gain excess profit. The level of predictability discovered here does not contradict market-efficiency theory.

This thesis investigates abnormal trading volume around scheduled and unscheduled announcements. The research is an extension of Chae (2005), Journal of Finance, Vol 60, which tests corporate announcements in the US stock market. In this thesis, Australian stocks are used to establish whether market characteristics affect trading behaviour around announcements. In addition, I extend the traditional methodology to overcome possible shortcomings in the previous studies. This thesis also discusses how information asymmetry affects the abnormal trading volume on the announcement day. In contrast to earlier studies, I nd abnormal trading volume does not change before either scheduled or unscheduled announcements, but, as expected, increases on and after the scheduled and unscheduled announcements. Information asymmetry increases trading volumes when unscheduled announcements are made, but has no effect for scheduled announcements. I show that the failure to adjust for the correlation between corporate events, results in abnormal trading volumes being detected prior to announcements. Differences between the Australian and US results can not all be explained by methodological differences. It appears that the underlying dynamics of the Australian market are different; casting doubts on the ability to generalize market characteristics from US based studies on abnormal trading volumes.

[Truncated abstract] This thesis examines and compares the performance of three classes of stock trading strategies in the Australian stock market from 1980 to 2002. ... The first segment of this thesis examines some simple technical trading rules with a twostep methodology ... Our standard test results show that technical trading rules generate excess returns higher than that of the buy-and-hold portfolio equivalent prior to 1991, but generate lower returns in the period post-1991. Bootstrap test results also show that addressing nonnormality, time-dependence and conditional heteroskedasticity in the data reverses the standard test outcome of predictability ... In addition, our sub-sample results also show technical trading rules becoming less profitable over time ... The second segment of this thesis examines trading rules based on the forecasts of four time-series models: the AR(1), AR(1)-GARCH(1,1), AR(1)-GARCH(1,1)-M and AR(1)- EGARCH(1,1) models. These time-series trading rules were examined with standard t-tests and found to be significantly less profitable compared to technical trading rules. Subsample results also show the time-series trading rules losing profitability over time, which supports the conjecture that the Australian stock market became increasingly efficient over time. The third segment of this thesis examines trading strategies based on various combinations of technical trading rules and time-series models ... Due to the weak performance of the time-series trading rules, our results show that combining technical rules with time-series models do not lead to improved forecast accuracy. Sub-sample results again show a strong decline in profitability post-1991, suggesting that technological advancements in the ASX since 1991 enhance market efficiency such that the above simple stock trading strategies are no longer profitable.

Consumption-based asset pricing models (CCAPMs) connect asset returns with consumption growth. The poor empirical performance of early consumption models has led to the development of a number of more sophisticated models. Nevertheless, most models focus on the US markets, and very few CCAPMs have been examined in the Australian context. Given the importance of CCAPMs, the purpose of this thesis is to examine the connections between asset returns in the Australian market and consumption variables. The thesis also extends the analysis to examine CCAPMs in an international setting. There are four essays in this thesis. The first essay undertakes a thorough investigation of the empirical support for consumption-based asset pricing models in the context of several major Australian asset classes. Using the generalised method of moments (GMM) econometric approach, my study begins with the classic CCAPM originally tested by Hansen and Singleton (1982, 1983). The empirical analysis is then extended to test more-recent specifications of the CCAPM, including the habit-formation models of Abel (1990) and Campbell and Cochrane (2000), and the time nonseparable model of Epstein and Zin (1991). For each of the models examined, the results provide cautious support for the CCAPM especially in relation to equity returns. Size-sorted portfolios (in particular, portfolios of small stocks) and fixed-income returns cause the CCAPM restriction to be rejected. It also presents results that raise questions over the benefits from extensions of the classic CCAPM, such as habit-persistence and recursive utility models. The second essay studies the empirical performance of a linearised version of the classic CCAPM in the Australian market. The studies of Faff and Oliver (1998) and Faff (1998) are extended by employing more recent data and examining 25 size/BM portfolios as well as industry portfolios. It is found that by using the lagged portfolio returns, the linearised CCAPM for both industry portfolios and 25 size/BM portfolios is generally not rejected. The third essay empirically examines conditional CCAPMs where the conditioning variables are consumption factors such as the consumption-wealth ratio proposed by Lettau and Ludvigson (2001a, 2001b), the surplus consumption ratio (Campbell and Cochrane, 1999), and the labour income to consumption ratio (Santos and Veronesi, 2006). Here long-horizon return predictability tests are conducted using these factors and cross-sectional tests on whether these factors are priced using both 25 Size/BM portfolios and industry portfolios. Utilising the Fama-MacBeth (1973) procedure, it is found that conditional models perform better than unconditional models. However, these conditional models do not outperform the Fama-French three-factor model. The fourth essay tests the world CCAPMs. Using data for 17 countries, the following are tested: the classic world CCAPM under the assumption of complete international markets integration, the heterogeneous world CCAPM under the framework of Constantinides and Duffie (1996) and the world habit models. The finding here is that a large risk aversion is needed to resolve the equity premium puzzle for the classic world CCAPM; however, adding a cross-country consumption dispersion factor into the model significantly lowers the coefficients of consumption risk aversion. Unconditional linear factor models are also studied where it is found that the world consumption growth and the dispersion of the cross-sectional consumption growth provide some explanatory basis for the variation in the cross section of excess stock returns. More sophisticated consumption models perform better than the classic world CCAPM. This thesis makes a worthwhile contribution to the research literature on CCAPMs in Australia which up to now has been limited. It performs out-of-sample tests of major CCAPMs utilising several Australian asset classes. It not only provides some insights into the return predictability of the aggregate market index in Australia, but also presents some evidence of the explanation of the cross section of stock returns using consumption variables. Further, this thesis adds to the understanding of the