주식 기대수익률 산출을 위한 빌딩블록 방법에 대한 연구

We conduct a study to examine the “Building Block” method for estimating stock returns, a widely used approach in practical finance. In our research, we compare its out-of-sample predictive power with historical returns and other traditional predictors for major investable indices. Our findings reveal that the Building Block method generally outperforms historical returns and other predictors like dividend yield and term spread. We notably discover that Building Block models including valuation changes demonstrate higher predictive power than those without. These results hold true for both monthly and yearly return predictions. They also remain robust when we alter indicators and the estimation period for valuation and earnings growth. The significance of our study lies in its academic verification of the Building Block method's predictive capabilities, suggesting its superiority over existing predictors. Our research provides valuable insights for institutional investors and Outsourced Chief Investment Officers(OCIOs), offering guidance for developing more sophisticated capital market assumptions crucial for strategic asset allocation.