Suppose that Y1, Y2, …, Yn constitute a random sample from a...

Suppose that $Y_{1}, Y_{2}, \ldots, Y_{n}$ constitute a random sample from a normal distribution with known mean $\mu$ and unknown variance $\sigma^{2}$. Find the most powerful $\alpha$ -level test of $H_{0}: \sigma^{2}=\sigma_{0}^{2}$ versus $H_{a}:$ $\sigma^{2}=\sigma_{1}^{2},$ where $\sigma_{1}^{2}>\sigma_{0}^{2} .$ Show that this test is equivalent to a $\chi^{2}$ test. Is the test uniformly most powerful for $H_{a}: \sigma^{2}>\sigma_{0}^{2} ?$

Key Concepts

Likelihood Ratio Test

The likelihood ratio test (LRT) is a fundamental approach in hypothesis testing used to compare the likelihoods under different hypotheses. It involves forming the ratio of the maximum likelihood under the alternative hypothesis to that under the null hypothesis, and then rejecting the null hypothesis if this ratio exceeds a certain critical value. The LRT is especially powerful in the context of simple versus simple hypotheses as established by the Neyman-Pearson lemma.

Neyman-Pearson Lemma

The Neyman-Pearson lemma is a central result in hypothesis testing which provides a method for finding the most powerful test for a given size when testing simple hypotheses. It states that the likelihood ratio test is the most powerful test for distinguishing between two simple hypotheses, meaning it maximizes the probability of detecting the alternative hypothesis when it is true, for a fixed significance level.

Chi-Squared Distribution

The chi-squared distribution arises naturally when dealing with the sum of squared independent standard normal variables. In the context of testing the variance of normally distributed data, the test statistic constructed from the sample variance follows a chi-squared distribution under the null hypothesis. This property allows the formulation of variance tests as chi-squared tests, facilitating the derivation of rejection regions using quantiles from the chi-squared distribution.

Uniformly Most Powerful (UMP) Test

A uniformly most powerful test is one that has the highest power against all possible alternatives in a specified class at a given significance level. In some hypothesis testing scenarios, such as when testing for a change in variance with a specific alternative value, the Neyman-Pearson lemma guarantees a most powerful test. However, when the alternative is composite (e.g., ?² > ??²), it is not always possible to have a UMP test, and the test developed for a specific alternative may not be uniformly optimal over all alternatives.

Transcript

Please give Ace some feedback