91Ó°ÊÓ

The chi-square distribution, often denoted as \(\chi^2(n)\), is a fundamental concept in statistics that plays a key role in hypothesis testing and confidence interval estimation for variance and standard deviation. It is a special case of the gamma distribution and is defined only for positive values. The chi-square distribution with \(n\) degrees of freedom has its applications in the chi-square tests for goodness of fit and independence.

• It is skewed to the right, with its shape depending on the degrees of freedom \(n\).
• The mean of the chi-square distribution is \(n\), while the variance is \(2n\).
• Understanding its properties is essential when engaging in statistical analyses that involve large sample sizes or variance estimates.

In the given exercise, \(Z_{n}\) represents a chi-square distributed variable with \(n\) degrees of freedom. As we look at the limiting distribution of \(W_{n}=Z_{n} / n^{2}\), it becomes imperative to understand how the chi-square distribution behaves as the degrees of freedom increase.

The Central Limit Theorem (CLT) is a powerful and widely used theorem in statistics, stating that the distribution of the sum (or average) of a large number of independent, identically distributed variables, as long as they have a finite mean and variance, will tend to be normally distributed, regardless of the underlying distribution. Key points to note include:

• The theorem provides a foundation for many statistical procedures, including confidence intervals and hypothesis tests.
• As the sample size increases, the distribution of the sample mean becomes increasingly normalized.
• The CLT is employed to approximate the sampling distribution of the mean for a set parameter, which is useful for prediction and analysis.

In the context of the problem we are considering, the variable \(Z_{n}\) has a chi-square distribution with finite mean and variance. Therefore, the Central Limit Theorem posits that for large values of \(n\), \(W_{n}\) would approach a normal distribution. However, since we are dealing with \(W_n = \frac{Z_n}{n^2}\) the situation is more nuanced due to the scaling by \(n^2\), which affects the limiting distribution.

A degenerate distribution is a probability distribution concentrated at a single point. In other words, it's a random variable with zero variance, meaning it has no uncertainty and takes a certain, fixed value with probability one. Notable aspects of this distribution include:

• Its probability mass function is equal to 1 at the point of concentration and 0 everywhere else.
• It doesn’t resemble common probability distributions like normal or uniform distributions, which spread out over a range of values.
• In practice, a degenerate distribution can represent a certain outcome in a random process.

As we see from our exercise, because the variance of \(W_{n}\) tends toward zero as \(n\) increases, ultimately, the distribution of \(W_{n}\) will converge to a degenerate distribution at 0. This makes intuitive sense because as we divide by a squared factor of the degrees of freedom, the chi-square distribution gets squeezed closer and closer to 0, hence the convergence to a certain value, not a range of values as we would expect with a typical distribution.