A measure of reproduction in human fecundability studies is the number of menstrual cycles required to achieve pregnancy which is assumed to follow a geometric distribution with parameter p. Tests of heterogeneity in the fecundability data through goodness of fit tests of the geometric distribution are developed, along with a likelihood ratio test statistic and a score test statistic. Simulations show both are liberal, and empirical level of the likelihood ratio statistic is larger than that of the score test statistic. A power comparison shows that the likelihood ratio test has a power advantage. A bootstrap p-value procedure using the likelihood ratio statistic is proposed.
Paul, Sudhir R.
"Testing Goodness Of Fit Of The Geometric Distribution: An Application To Human Fecundability Data,"
Journal of Modern Applied Statistical Methods:
2, Article 8.
Available at: http://digitalcommons.wayne.edu/jmasm/vol4/iss2/8