Question 1026440: This applet carries out tests at a fixed significance level.
When you arrive, the applet is set for the cola tasting test of Example 18.6.
That is, the hypotheses are
H0 : μ = 0
Ha : μ > 0
We have an SRS of size 10 from a Normal population with standard deviation σ = 1, and we will do a test at level α = 0.05.
At the bottom of the screen, a button allows you to choose a value of the mean μ and then to generate samples from a population with that mean.
Step 1:
Set μ = 0, so that the null hypothesis is true.
Each time you click the button, a new sample appears.
If the sample x lands in the colored region, that sample rejects H0 at the 5% level.
Click 100 times rapidly, keeping track of how many samples reject H0.
Use your results to estimate the probability of a Type I error.
If you kept clicking forever, what probability would you get?
Give your answer to 2 decimal places.
Step 2:
Now set μ = 0.8. Example 18.6 shows that the test has power 0.812 against this alternative. Click 100 times rapidly, keeping track of how many samples fail to reject P 0. Use your results to estimate the probability of a Type II error. If you kept clicking forever, what probability would you get?
Answer by ikleyn(52787)   (Show Source):
You can put this solution on YOUR website! .
What is this ???
Why it is here ???
What we should do ???????
As I understand, somebody copy, past and submit without thinking.
|
|
|
