The chi-square test is a well-known goodness-of-fit test. It is available for arbitrary alternative hypothesis, particularly for a very general alternative. But when the alternative is an "one-sided" hypothesis, which usually appears in genetic linkage analysis, the chi-square test does not use the information offered by the one-sided hypothesis. Therefore, it is possible that an appropriate one-sided test, which uses the information, will be better than the chi-square test, this paper gives such an efficient one-sided test. The Monte Carlo Simulation results show that it is more powerful than the chi-square test, and its power is over two times greater than that of the chi-square test in most situations and over six times greater in the best case.
|