| This dissertation examines the optimal asset allocation between the two main financial asset categories, stocks and bonds, for individual investors maximizing a power utility function over wealth in an inter-temporal framework. The present work is a step forward by relaxing certain restrictive assumptions consistent with recent literature, and showing empirically that the risk-return profile of financial assets changes with the holding period. With respect to stocks, we show that while the expected annualized returns grow proportional to time, the variance grows as well, but at a slower rate. The investors' optimal weight of the risky asset depends precisely on this ratio of expected returns to variance, thus in the long run the allocation to stocks should increase. We implement this analysis in a regime of a conventional risk-free asset, as well as considering the uncertainty associated with the historical T-bill returns.{09}Statistically we employ the re-sampling procedure of bootstrapping to eliminate the structure in data generated by overlapping returns. |
