Null hypothesis and alternative hypothesis

175 Views Asked by At

A statistical variable X can have only one of the two following distributions:

H0: f0(x) = 0.2-0.02x 
H1: f1(x) = 0.02x

In both cases, the support of the statistical variable is:

0 <= X <= 10

The decision rule for to test the hypothesis, based on a single casual observation, consider the following rejection region for H0:

Rif = {X >= 8}

Is is required to:

a) Evaluate the significance level (alpha);

b) Evaluate the power of the test (1-beta);

My partial solution: I have written that alpha is the probability to reject H0 when H0 is true (error of first type). 1-beta is the probability to reject H0 when H0 is false (this means, the probability to make the correct decision).

How can I continue to solve?

Thank you very much for considering my request.

1

There are 1 best solutions below

7
On

If $H_0$ is true, then $X$ has the probability density $f_0(x) = 0.2 - 0.02x$ for $0 \le x \le 10$. So what is $\Pr[X \ge 8]$ under this assumption? It is $$\Pr[X \ge 8] = \int_{x=8}^{10} f_0(x) \, dx = \ldots?$$

Similarly, under the assumption that $H_1$ is true, what is the probability $\Pr[X \ge 8]$?