Applied Modelling and Data Analysis

Applied Modelling and Data Analysis

1.INTRODUCTION

Equity returns change over time. These changes might be due to economy wide factors or to firm-specific factors, such as changes in accounting variables like an increase in leverage, earnings/price and book-to-market equity.

Asset prices are commonly believed to react sensitively to economic news. Daily experience seems to support the view that individual asset prices are influenced by a wide variety of economic events and that some events have more pervasive effect on asset prices than do others.

Consistent with the ability of investors to diversify, modern financial theory has focused on pervasive or systematic influences as the likely source of investment risk. The relation between the stock market and macroeconomic forces has been widely analyzed in the finance and macroeconomic literature. The linkages between equity prices and variables such as money supply, inflation and industrial production are of crucial importance not only in analyzing equity returns, but also in understanding the connections between expected returns and the real economy.

Ross (1976) with the arbitrage pricing theory (APT) introduced the idea that few factors govern the asset prices while the work of Chen, Roll and Ross (1986b) has given new impetus to research on the macroeconomic determinants of equity returns. Research has concentrated mainly on the significance of the risk premia attached to each macroeconomic factor, providing considerable evidence that state variables such as industrial production growth, default risk premia and yield spreads between long and short-term government bonds are important in explaining equilibrium asset prices. Since the theory does not indicate which macroeconomic factors to include on the model, statistical methods are applied to infer the presence of factors from patterns in the time series data on assets' rates of return. The specialised tools of factor analysis and principal components analysis are employed in these investigations.

This paper identifies the macroeconomic state variables that influence Greek equity returns for the period from 1997 to 2004. Since different economies are likely to be idiosyncratic to some degree in selecting the relevant macroeconomic factors the attention is not confined to variables used in previous research on other countries, but rather to variables which might have a specific relevance for the Greek stock market. The study examines the ability of the macroeconomic series to explain their relationship with the Greek stock market once based on their ability to predict the factor scores estimated using Factor Analysis and once by regressing directly the macroeconomic variables in examining their significance in explaining expected stock returns.

Assuming that asset prices depend on their exposure to state variables in the economy, how well does factor analysis illustrate this exposure? How well do parameter estimates for innovations in pre-specified macroeconomic variables describe security returns? What are the implications of the answers to these questions for tests of the Arbitrage Pricing Theory (APT)? These three questions are addressed in the following pages.

Despite the extensive literature on testing the relation between macroeconomic variables and stock market returns very few studies has been conducted in regard to the ...