Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $X_1, \dots, X_n$be a random sample of size $n$ from the continuous distribution with pdf given by

$f_x(x|\theta) = \theta x^{\theta - 1} I(x)_{(0,1)} I(\theta)$.

1) Show that $T = \prod^n_{i=1}X_i$ is sufficient for $\theta$.

2) Let $0 < \theta_0 < \theta_1$. Show that $\frac{f_x(x|\theta_1)}{f_x(x|\theta_0)}$ is an increasing function of $T$.

3) We are testing $H_0: \theta\geq 4$ vs. $H_A: \theta < 4$. Let $W = -\ln(T)$. Show that the test rejects $H_0$ if and only if $W>k$ is a UMP level $\alpha$ test where $\alpha = P(W>k|\theta = 4)$.

4) Derive the distribution for $W$.

5) Suppose that $n=14$. Use the appropriate tables to determine the level $\alpha$ of the UMP test from 3 when $k =4.0775$

1) Seems not too bad using the Factorization Theorem and setting that to be T and for 2), this seems live a lot will cancel and give me what I want when I plug in the quotient. I am pretty stuck with 3) and how to set that up but I think it is a form of a Beta distribution and I know I need to use Neyman-Pearson, I am just not quite sure how. 4) I think I have this by doing a transformation and getting an Exponential distribution, but not totally sure that is correct. 5) Not sure if this will be simpler once I get the others done or not.

Any help is greatly appreciated.

share|cite|improve this question

I'm not sure what I have is correct, so I won't put down what I got so that I won't confuse you, when I'm already confused, but I will make a note for your part 3, which I'm sure is correct.

For part 3, what did you do to get the Beta distribution?

$\bf{Note:}$ The Neyman-Pearson Lemma can only be applied in a simple hypothesis test. Are your $H_{o}$ and $H_{A}$ simple hypotheses? If it's not a set of simple hypotheses, then is there another theorem that you can use that uses the fact that T is a sufficient statistic and that the MLR (from part 2) is non-decreasing? If you answered yes, then which theorem is it?

It's the Karlin-Rubin Theorem!

Here's where I found information on the Lemma. Lemma

For part 4, I also got the distribution is exponential, but then like I said before, I'm also not sure if I'm right on this question.

Good Luck with Part 5, I'm unsure as well.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.