Do Economic and Technical Variables Matter on Equity Premium Forecasting?: An Evidence of the Korean Stock Markets

Stock return forecasting is a long-lasting issue in the field of finance. This study investigates on the equity premium predictability of economic and technical indicators in the Korean stock markets. To this end, we select a set of 33 technical and 14 economic variables for KOSPI 200 and KOSDAQ index and extract the principal components of these variables by principal component analysis (PCA). Then, based on a standard predictive regression framework, the variables and principal components’ predictability on the equity premium and the investor sentiment are examined. Also, we test the relationship between investor sentiment and the equity premium. The major results of this study are as follow: First, economic variables have a reliable predictive power on equity premium compared to technical variables. Technical indicators have an explanatory power during economic expansions, and a considerable improvement in explanatory power is achieved by controlling the economic conditions and combining the principal components extracted from the two predictor groups. Second, technical indicators better detect the changes in the sentiment of investors than the economic variables do. The predictive power of technical indicators of investor sentiment is more robust in the KOSDAQ market than in the KOSPI 200 market, regardless of economic conditions. Third, in the KOSPI market, where institutional investors have a relatively high share of trading, investors appear to consider all three types of technical indicators in their investment decisions. On the other hand, in the KOSDAQ market, where the proportion of individual investors is high, investors tend to rely on price indicators.