I'm currently reading about statistics and I'm trying to figure out all the different distributions, tests, hypotheses etc.
I have come across an exercise which is basically like this:
I'm given four numbers including x1, x2, x3 and n, where n is the total number of trials. The numbers for x1, x2 and x3 are all different.
Then I'm given a model M0 which states that (X1, X2, X3) is distributed as the multinomial distribution with parameters n and (π1, π2, π3) where (π1, π2, π3) belongs to ∏^(3).
My task is show that there is not equal probability for the three events.
So, my first thought was to set up a hypothesis H0 which should state something like (π1, π2, π3) = (p1, p2, p3), p1 ≠ p2 ≠ p3, p1 + p2 + p3 = 1.
I don't know if this would be the correct way to set up the hypothesis?
Also, I'm confused about what test should be used to show that the probability isn't equal for the three events. Clearly, it isn't a t-test or an F-test, since these are for mean and standard deviation/variance. Could it be a multinomial test?
Help would really be appreciated.
Nobody that can help?