A Structured Financial Statement Analysis and the Direct Prediction of Stock Prices in Korea

Using a large sample of firms listed on the Korea Stock Exchange, this paper evaluates the usefulness of a structured, programmable financial statement analysis for investment decisions. In doing so, we develop a firm valuation model which links a firm's market value with fundamental variables such as the ability of a firm to generate cash flows, growth potentials, and risk. We predict a firm's intrinsic value directly from an extensive set of financial statement variables which proxy for the theoretical variables implied by the model. We then construct a series of trading strategies with zero net investment (called D-strategies) on the basis of D-values which measure percentage differences between predicted intrinsic values and observed market values. We observe that the market-adjusted and size-adjusted (hedge-portfolio) returns to the most conservative D-strategy turn out to be in the order of 16.92% and 11.44%, respectively,for the 12-month holding period. When our sample is stratified into two sub-samples based on firm size, the D-strategy yields higher excess return for the small-firm sub-sample than for the large-firm sub-sample. The above evidence, taken as a whole, strongly indicates that one can construct a profitable trading strategy by directly predicting intrinsic values through a structured financial statement analysis such as ours.