A new empirical approach to explain the stock market yield: a combination of dynamic panel estimation and factor analysis

This paper presents an empirical approach that combines competing paradigms of mod-eling in empirical capital market research. The approach simultaneously estimates the explanatory power of fundamentals, expectations, and historic yield patterns, making it possible to test the extent to which the efficient market hypothesis, fundamental data analysis, and behavioral finance contribute to explaining stock market yield. The core of the approach is a dynamic panel model (Arellano-Bond estimator with an MA restric-tion of the residuals), complemented with an upstream factor analysis to reduce multi-collinearity. Due to the complexity of the data set, a great many parameters that influ-ence the yield can be determined. Highly significant parameter estimates are possible even though the information in the data set is interdependent. For the German stock market (the 160 companies listed in DAX, MDAX, SDAX, and TecDAX), the quarterly yield is analyzed for the period between 2004 and 2009. The model has high explanato-ry power for the entire observation period, even in light of the fact that the period in-cludes the financial crisis of 2008.