I am struggling with finding a fitting distribution for a chi square goodness-of-fit test in R. My quantile is 66.77, but my test statistic is staying far apart from it. The best distribution I found is an exponential distribution with parameter 0.143, whereby my test statistic has a value of 109.11. I also tried an erlang distribution, but no parameters were better than exponential parameters. Which distributions could also work, and which formulas do I have to use to calculate the expected frequencies for them? I also tried to use the provided chi square test in R, but I get very different test statistics compared to the test statistics I calculated with a for loop by the sum of (observedValue - expectedValue)^2/expectedValue.
Asked
Active
Viewed 12 times
0
-
This is really a statistical question rather than a programming question. – user2554330 May 31 '23 at 08:46