ADM 2304 Lecture Notes - Lecture 12: Null Hypothesis, Analysis Of Variance, Random Variable

• School: University of Ottawa
• Department: Administration
• Course: ADM 2304
• Professor: Tony Quon

Document Summary

Again, we will use tables to determine if our f-statistic is large enough to reject at the pre-speci ed signi cance level. > again, the null hypothesis is that there is no effect. > if the means from the different treatment groups are different enough we conclude that the treatment did in fact have an effect. > the aim is to show a relationship between treatment and effect. At a means are equal least one mean differs from others the. > for one-way anova, we assume that the response, x, will be a random variable centered around some mean. > for i = 1, ,i (the populations) and j = 1, . So xij is the jth observation from the ith population (or treatment). If the null hypothesis is true then > ijx ij. Mst = sst/(i 1) where i is the number of populations. Mse = sse/(n - i) where n is the sum of all sample sizes.