Please help with simple statistics problem

  • Thread starter Thread starter bryandnk
  • Start date Start date
  • Tags Tags
    Statistics
Click For Summary
The discussion revolves around a statistics problem involving two normal distributions, N(mu1, 400) and N(mu2, 225), and the hypothesis testing of theta = mu1 - mu2. The user attempts to derive the sample size n and critical value C for a power function pi(theta=10) = 0.95 at a significance level of alpha = 0.05. They calculate n to be approximately 67.65 and round it up to 68, noting that increasing the sample size enhances the test's power. The calculations are confirmed as correct, emphasizing the importance of rounding up to ensure adequate power in hypothesis testing.
bryandnk
Messages
4
Reaction score
0

Homework Statement


Consider the distributions N(mu1, 400) and N(mu2, 225). Let theta = mu1-mu2 and x and y be the observed means of two independent random samples, each of size n, from these two disbtibutions. We reject H(0) : theta = 0 and accept H(a): theta >0 if and only if x-y >=C. If pi(theta) is the power function of this test, find n and C so that pi(theta=10) = 0.95 at significance level alpha = 0.05.

Thank for anyone's help.


The Attempt at a Solution


<br /> <br /> {X-Y-10\over\sqrt{{400\over n}+{225\over n}}}=-1.645<br />

and:

C = X-Y=1.645\sqrt{{400\over n}+{225\over n}}


So if I subtract the first equation from the 2nd, I get:

10=3.29\sqrt{{400\over n}+{225\over n}}

and n = 67.65, but since it should be a whole number, we round up to 68?

Is any of this right, or what's the correct answer?
 
Physics news on Phys.org
no one can help?
 
So the first equation is from the true distribution of x - y under H(a) which is Normal(10, 625/n) due to independence.

The second equation comes form the upper tail test of x - y, which only rejects when the test statistic is greater than (in this case) 1.645.

Seems to me you've done this correctly; you want to round up to 68 because increasing sample size will only increase power.
 
First, I tried to show that ##f_n## converges uniformly on ##[0,2\pi]##, which is true since ##f_n \rightarrow 0## for ##n \rightarrow \infty## and ##\sigma_n=\mathrm{sup}\left| \frac{\sin\left(\frac{n^2}{n+\frac 15}x\right)}{n^{x^2-3x+3}} \right| \leq \frac{1}{|n^{x^2-3x+3}|} \leq \frac{1}{n^{\frac 34}}\rightarrow 0##. I can't use neither Leibnitz's test nor Abel's test. For Dirichlet's test I would need to show, that ##\sin\left(\frac{n^2}{n+\frac 15}x \right)## has partialy bounded sums...