미래순이익예측모형에 관한 연구: 횡단면적 모형과 재무 분석가 예측 비교를 중심으로

초록

We utilize the cross-sectional earnings model to predict future earnings of firms in the Korean capital market. Following Hou et al. (2012), We generate one- and two-year-ahead future earnings from the cross-sectional model using a few of accounting variables to forecast future earnings for Korean firms. The sample used in this study consists of non-financial public firms in the Korean stock market from 2002 to 2010. We evaluate the performance of the model-based forecasts vs. analysts’ forecasts in terms of accuracy and bias. We find that the model-based forecasts are less accurate than analysts forecasts but they are less biased than analysts forecasts for one-year horizon. For two-year horizon, the model-based forecasts are more accurate and less biased than analysts’ forecasts. We also find a higher level of earnings response coefficient for the model-based forecasts than analysts forecasts in the association framework. This empirical evidence suggests that the model-based earnings forecasts are useful proxies for market expectations of future earnings. Overall, this study suggests that earnings forecasts from the cross-sectional model can be useful in the Korean capital market where analyst coverage is limited.

키워드