University of Washington - Department of Statistics
A common problem with financial historical data is that they often have unequal lengths of histories. Examples include country market indices, currency rates and hedge fund returns histories. Practitioners often deal with such issues by truncating all the series so that the remaining data have the same length, which is apparently not an ideal solution. We discuss existing statistical methods that utilize the full data set, such as maximum likelihood estimation and multiple imputation. We also propose a few new nonparametric or semi-parametric methods, most notably the bootstrap method and the backfill method. The performance of different methods are investigated through asymptotic theories, as well as Monte Carlo simulations. Our results show that, all combined-sample methods are more efficient than the truncation method, while different combined-sample methods may outperform the rest under different settings.