Nonparametric tests in testing composite hypotheses on goodness of fit to exponential family distributions
B.Yu. Lemeshko and A.A. Maklakov
Novosibirsk
Pages: 3–18
Abstract
Statistical modeling methods are used to investigate statistic distributions of nonparametric Kolmogorov, and also and Mises tests for composite hypotheses on goodness of fit of empirical data to the exponential family distributions. Models of limiting distributions of statistics of the considered goodness of fit tests are constructed for different values of the parameter of form of exponential family and are recommended for application to problems of statistical analysis.