(Solved): Problem 1: Deriving the Lower-Tailed Hypothesis Test Consider testing the set of hypothesis [H0: ...
Problem 1: Deriving the Lower-Tailed Hypothesis Test Consider testing the set of hypothesis [H0?:HA?:?p=p0?p<p0?? at an arbitrary ? level of significance. Define the test statistic TS to be TS=np0?(1?p0?)??p??p0?? a. Show that TS?H?N(0,1). If your answer depends on a set of conditions to be true, explicitly state those conditions. b. Argue, in words, that the test should be of the form decision(TS) { reject H0? fail to reject H0?? if TS <c otherwise ? for some constant c. As a hint, look up the logic we used in Lecture 13 to derive the twotailed test, and think in terms of statements like " p^? is far away from p0? ". You do not have to find the value of c in this part. c. Now, argue that c must be the ?th percentile of the distribution of the standard normal distribution (not scaled by negative 1), thereby showing that the full test takes the form decision(TS) ={ reject H0? fail to reject H0?? if TS <z?? otherwise ? where z?? denotes the (?)×100th percentile of the standard normal distribution.
I Result: Upper-Tailed Test When testing the hypotheses [H0?:HA?:?p=p0?p>p0?? at an ? level of significance, the test takes the form decision(TS) ={ reject H0? fail to reject H0?? if TS >z1??? otherwise ? where: - TS=np0?(1?p0?)??p^??p0?? - z1??? denotes the (1??)×100th percentile of the standard normal distribution. provided that - np0??10 - n(1?p0?)?10
Sure, I can help you with that.
**a. Show that TS^ underline H 9 mathcal N(0, 1) . If your answer depends on a set of conditions to be true, explicitly state those conditions.**
The sampling distribution of the test statistic TS is given by:
```
```
This is a standard normal distribution with mean 0 and standard deviation .
The conditions for this to be true are:
* The sample size n must be large enough.
* The population must be normally distributed.
* The sample must be randomly selected from the population.
**b. Argue, in words, that the test should be of the form decision(TS) = (reject Ho fail to reject Ho otherwise for some constant c. As a hint, look up the logic we used in Lecture 13 to derive the two- tailed test, and think in terms of statements like "p is far away from p_{0} ^ prime prime . You do not have to find the value of c in this part.**
The test should be of the form decision(TS) = (reject Ho fail to reject Ho otherwise if the test statistic TS is far away from the hypothesized value p0. The value of c is the critical value that separates the rejection region from the non-rejection region.
The logic for this is similar to the logic used in Lecture 13 to derive the two-tailed test. In that case, we used the fact that the p-value is the probability of obtaining a test statistic at least as extreme as the one observed, under the null hypothesis. We then set a significance level alpha, and rejected the null hypothesis if the p-value was less than alpha.
In this case, we are using a one-tailed test, so we only need to consider the probability of obtaining a test statistic that is less than the hypothesized value p0. We can set a significance level alpha, and reject the null hypothesis if the test statistic is less than the critical value c, which is the (1-alpha)th percentile of the standard normal distribution.