Orthodox, established methods of mathematical statistical science are used to assess the trustworthiness of each of the programs that are presented here. Peer-reviewed articles by academicians, such as on stock market momentum, incorporate tests of significance and provide starting points. But additional, out-of-sample testing and other hypothesis testing is also conducted.
The Traded Portfolio project conducts out-of-sample testing in the time domain in order to arrive at an additional unbiased characterization of any given asset allocation rule’s likely future performance. Currently “p values” are computed using Monte Carlo permutation, in formal refutation of the null hypothesis. Cross validations are performed and confidence intervals pertaining to projected rates of return are computed.
Every offering presented here has been crafted based on risk-adjusted returns, and every effort has been made to find strategy specifications that minimize downside risk.
The writeups of the Notes menu provide additional details and go over some of the opportunities and complications.