STAT230 Study Guide - Midterm Guide: Maximum Likelihood Estimation, Confidence Interval, Sampling Distribution

• School: University of Waterloo
• Department: Statistics
• Course: STAT230
• Professor: Don Mc Leish

Document Summary

In the data stage, we assume the data we will measure follows a distribution. Using mle, we obtain estimates for our parameters such as. But if we have yet to collect any data, we replace the data points with random variables, so now we have a corresponding estimator (a random variable) such n. 1 n i as is now a random variable, it also has a distribution. The sampling distribution obviously is related to the population distribution because the estimator is a function of the random variables defined by our data. Using sampling distributions, we can now make probability statements about the accuracy of our estimates for parameters. T where n is the degrees of freedom. S and t and s are independent then. T, a random variable following a mt distribution is defined as. We say mtt ~ where m is the degrees of freedom. Properties of mt distribution: the t distribution is symmetric about zero, thus.