STAT 2040 Lecture Notes - Lecture 10: Paired Difference Test, Statistical Unit, Variance
24 views4 pages
1 Apr 2015
School
Department
Course
Professor
Document Summary
Use confidence intervals estimate the true difference in means of 2 groups: the mean of the sampling distribution of is: is an unbiased estimator of, since the 2 samples are independent var", standard deviation. *the sampling distribution of distributed populations (regardless of the sample sizes). will be exactly normal if we are sampling from normally. *if sampling from non-normal distr" w. big sample size, central limit theorem the sampling distribution of will be approximately normal. Hypothesis tests and confidence intervals for two independent samples (when are known) z procedure. When are unknown t procedures (pooled & welch) An exact procedure that requires the additional assumption that. The conservative estimate of the degrees of freedom is the lesser of n1 1 and n2 1. Independent simple random samples: normally distributed populations. (or lrg sample size, equal population variances. Upside: if the two pop have equal variances, then the test statistic has exactly a t-distribution (not just an approximation).