MAT 117 Chapter Notes - Chapter 12: Null Hypothesis, Contingency Table, Test Statistic
Document Summary
Goodness-of-fit test for a multinomial experiment: a multinomial. Nov. 14th, 2016 experiment consists of a series of n independent and identical trials, such that for each trial. The probability pi associated with the ith category remains the same. The sum of the probability is one, that is p(cid:883) + p(cid:884) + + pn = (cid:883). When p = 2, it is a binomial experiment. Test statistic for goodness-of-fit test: for a multinomial experiment with k categories, the value of the test statistic is calculated as. Where df = k 1, and oi is the observed frequency and e = npi is the expected frequency in the ith category, respectively and there are 5 or more categories. Important: the test is valid when the expected frequencies for each category are five or more. As we run the hypothesis testing for goodness of fit, it is important that we state the null hypothesis h0 and alternative hypothesis h1.