91Ó°ÊÓ

The moment generating function, commonly known as mgf, is a powerful tool in probability theory. It's a function that helps to characterize all the moments of a random variable. In simple terms, it acts like a "fingerprint" for the distribution.
For an exponential distribution with a rate parameter of 1, the mgf is given by \( M_X(t) = \int_0^{\infty} e^{tx} e^{-x} dx \). Simplifying it, you get \( M_X(t) = \frac{1}{1-t} \), valid for \( t < 1 \). The mgf provides us with easy access to obtain moments by simply differentiating the function with respect to \( t \).
When dealing with sample means, the mgf for \( \bar{X}_n \), the mean of a sample of \( n \) independent and identically distributed exponential random variables, is \((M_X(t/n))^n\). It gives \((1 - t/n)^{-n}\), because the mgf of each random variable is scaled by \(1/n\) for taking the average of the whole sample.

The limiting distribution describes the behavior of a sequence of random variables as the number of samples, \( n \), approaches infinity. It's a crucial concept in probability and statistics because it allows us to understand what happens to sample-based measures when the sample size is very large.
In this exercise, you see the role of limiting distribution when computing \( Y_n = \sqrt{n}(\bar{X}_n - 1) \). For large \( n \), the mgf of \( Y_n \) approaches the mgf of another well-known distribution. By evaluating the behavior of \( M_{Y_n}(t) \) as \( n \) goes to infinity, we find that it converges to a specific form, indicating the distribution \( Y_n \) starts to resemble.
Thus, the limiting distribution essentially forms the foundation for the Central Limit Theorem, which states that sample means will tend toward a normal distribution, regardless of the original distribution of the data, given a sufficiently large \( n \).

The standard normal distribution is a probability distribution with a mean of 0 and a standard deviation of 1. This is denoted as \( N(0,1) \).
It's a cornerstone of statistics because it simplifies the process of computing probabilities and estimating parameters. When random variables are converted to this format through **standardization**, comparisons are made easier across different datasets.
Within this particular problem, as \( n \) approaches infinity, the random variable \( Y_n \) converges to a standard normal distribution. Essentially, the expression \( \sqrt{n}(\bar{X}_n - 1) \) symbolizes how far the sample mean is from the true mean in standardized form. It's this standardization which makes it culminate into \( N(0,1) \) when \( n \) is large enough, by virtue of the Central Limit Theorem.