Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I have some random variable: ${X ~ N(\mu, 1)}$

I define null hypothesis and alternative hypothesis as follows:

${H_0 : \mu = 0}$

${H_1 : \mu \neq 0}$

Then, let ${x}$ be a number that comes out of my random variable.

I want to reject my null hypothesis if ${|x| \geq 3}$

And then I need to calculate probability of possible errors.

So, first of all I need to find the test statistic:

${ \displaystyle z = \frac{\bar{x} - \mu_0}{ \displaystyle \frac{s_x}{\sqrt{n}}}}$

I know that ${\mu_0}$ = 0 and sample size should probably be equal to ${1}$ (?).

How to calculate ${\bar{x}}$ (sample average)? Is it the same as expectation?

My question may be trivial, but I'm totally confused with it.

share|improve this question

1 Answer 1

up vote 1 down vote accepted

A "sample" of $1$ is highly unusual, and therefore it is reasonable to be confused. The random variable is known to have variance $1$. The mean is not known, else we would not be doing the test of hypothesis.

As you wrote, we reject if the absolute value of the sample mean is $\ge 3$. The sample mean in this case is just the value taken on by the random variable. (The average height of a group consisting of one person is the height of that person.)

Under the hypothesis $H_0$, the absolute value is $\ge 3$ with probability $\Pr(|Z|\ge 3)$, where $Z$ is standard normal. So you will falsely reject $H_0$ with probability $\Pr(|Z|\ge 3)$.

You can calculate this number from information available in tables of the standard normal. I get about $0.0026$.

share|improve this answer
    
Thank you very much! –  Edward Ruchevits Nov 15 '12 at 23:11

Your Answer

 
discard

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.