I have the following problem:

Consider two probability density functions on $[0,1]: f_0(x) = 1$, and $f_1(x) = 2x$. We want to test such hypotheses: $H_0: X \sim f_0(x)$ versus the alternative $X \sim f_1(x)$, with significance level $\alpha = 0.1$, how large can the power possibly be?

My approach:

Let $\{\Gamma_0, \Gamma_1\}$ be a partition of $[0,1]$ into sets and the decision rule be that hypothesis $H_i$ is true if the observation $X$ belongs to $\Gamma_i$. Then, the probability of type 1 error is $$\alpha = \int_{\Gamma_1} f_0(x)\,\mathrm dx = 0.1$$ while the power of the test is $$power = 1 - \beta = \int_{\Gamma_1} f_1(x)\,\mathrm dx.$$ Now,since $f_0(x)$ is the uniform density, we know that the total "length" of the set $\Gamma_1$ is $0.1$

Then looking at the graph of density $f_1(x)$ i have realized, tha in order to maximize the power of the test i should take the region from 0.9 to 1.

So the maximal power of the test is :

$$power = 1 - \beta = \int_{0.9}^{1} 2x\,\mathrm dx. = 0.19$$

Is my solution correct? And are there some other approaches for solving this exercise? I have thought about Likelihood-ratio test but i don't know how to apply it to my problem.


1 Answer 1


Your answer looks correct to me.

The likelihood ratio test would have the rejection rule for the form reject if $\frac{L_0}{L_1} \leq c$, where $L_0$ and $L_1$ are the likelihood under the null and alternative.

So in your problem the rejection rule becomes: "reject if $\frac{1}{2x} \leq c$".

i.e. reject if $x \geq k$ (for some $k$)

Now the $k$ required to get $\alpha = 0.1$ is $k=0.9$.

So in this case the LRT is the same as your test.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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