STAT 1100 Chapter Notes - Chapter 13: Test Statistic, Random Variable, Null Hypothesis
Document Summary
Let suppose that we have a sample of several observations on a given random phenomenon defined by a random variable whose probability distribution is not precisely known. This fact prevents the use of parametric tests studied in previous chapters because these tests require knowledge of the probability distribution of the studied phenomenon. The usefulness of goodness-of-fit tests is to compare the distribution of the random variable (observed distribution) with known theoretical distribution (normal distribution, fish distribution, student distribution,) that we believe it could be the distribution of . If the results of the goodness-of-fit test involve the fit of the distribution observed with a given theoretical distribution, it will allow to better understand and control the studied and also to perform some parametric tests. To perform the goodness-of-fit tests, we must begin to determine the theoretical distribution that we assume to be the distribution of the studied phenomenon.