City Research Online

Abstract

There is a lot of evidence in the academic literature that stock returns are predictable especially when various fundamental variables are used as predictors. Firm - specific variables, such as market value, book - to - price, cash flow - to - price, earnings yield or earnings growth have been extensively used either directly in static asset pricing models, as instruments in a conditional pricing framework, or as important factors in the construction of investment strategies. Whether the relationship between firm-specific variables reflects compensation for risk or some sort of market inefficiency is not yet clear. This thesis examines the characteristics and performance of various investment strategies constructed using fundamental variables, and investigate whether the difference in returns between portfolios can be attributed to differences in sensitivity to market industry and macroeconomic risk factors. Furthermore, market overreaction to past growth and analysts’ earnings projections is examined as alternative explanation for existence of the value-growth premium.

The second part of the thesis focuses on the short-term return variability of portfolio strategies constructed using market value and book-to-price variables. The volatility of the size and value premiums suggests that a consistent bet to a certain investment philosophy or style might not be the ideal investment choice. The thesis explores the feasibility of style rotation strategies after taking account different levels of forecasting skill and transaction costs and tests out-of-sample a model that utilises mainly macroeconomic factors to predict the next months’ size and value premium. Finally, the thesis analyse the volatility characteristics of style portfolios and proposes volatility specifications to predict future variances at different horizons.

Publication Type:	Thesis (Doctoral)
Subjects:	H Social Sciences > HG Finance
Departments:	Bayes Business School > Bayes Business School Doctoral Theses Bayes Business School > Finance Doctoral Theses

Export

Downloads