| Capitalization weighted indices have been widely used till now. However, the "return drag" in this kind of index occurs because cap-weights are too large or small for overpriced or underpriced stocks. Under noisy market hypothesis, capitalization weighted indexation is proved to be suboptimal. Since2005when Arnott first proposed new market indices based on firms’ fundamental size estimated by accounting metrics, the so-called fundamental indexation has been widely studied and tested in different markets in order to find an indexation better off the conventional cap-weighted indices and substitute it as a more appropriate approximation of "market portfolio". Basically, constitution for fundamental indices is under the thought of estimating the fundamental weights instead of fundamental value.In our study, based on theoretical support, we conduct9sectors of fundamental-weighted indexation in China A-share market as well as market capitalization-weighted sector indices. One is to pick and weight stocks for sector indices according to accounting metrics, namely book value, operating income, cash flow, and cash dividend for all A share. The other is by free-float market capitalization weights of all sectors for all A share. We evaluate fundamental indices by comparison with the conventional capitalization weighted indices in profitability, mean-variance efficiency and robustness and discuss the source of excess return if there is any in fundamental indices by testing in Fama-French three factor model.Our empirical study results show that in Chinese A-share market, fundamental sector indices via accounting metrics outpace capitalization weighted sector indices in our testing period from2004to2013with better performance; however, since the volatility of most fundamental indices via accounting metrics are much more bigger than that of cap-weights, this kind of fundamental indexation may work for China, but not as those in the US. |
PDF Full Text Request