How can i calculate critical values of Goodness of fit Tests for Weibull Distribution using pearson system via Mathcad program?
The goodness-of-fit statistic is equal to B = SIGMA(O(I)-E(I))^2/(E(I))^2 and follows a chi-square distribution with n-1 degrees-of-freedom. The E(I)'s are expected frequencies obtained from the hypothesized Weibull distribution whereas the O(I)'s are the observed frequencies. The critical value of B for a significance level of alpha is the chi-square alpha value.
Ette - How are the frequencies defined? Does this require constructing categories? Is this related to the one-sample Kolmogorov-Smirnov test?
David
You partition the data space into classes. Observe the frequencies O(I) and calculate the respective expected frequencies E(I). For example if the data range from 1 to 30, you can have the classes like 0-4, 5-10, 11-14, 15-20, 21-24, 25-30 which is a partition of the data space. Observe the frequencies O(I) and calculate the respective expected frequencies E(I). Yes, Kolmogorov_Smirnov one-sample goodness-of-fit test may be used also.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics