STA-3024 Lecture Notes - Lecture 19: Standard Deviation, Decision Rule, Null Hypothesis
Document Summary
Set hypothesis set signiicance level collect data decision rule: p-value a measure of risk, if the p-value is low, the null must go , reject null: there is enough evidence to conclude that ______ The null hypothesis is rejected when the actual value of interest is either less then or greater than the hypothesized value: 1 standard deviaion distance requires a lot of data, 3 standard deviaion distance requires not much data. Ttest procedure for one sample: proc test data= Tests and conidence intervals produced in proc test using: sides= u for upper tail tests and cls, sides=l for lower tail tests and cls, default in sas is a two-tailed test, h0: u1=5. Two sample: test score data set: h0: u1=u2, stacked vs unstacked data sets. F-test for equality of variances: assesses whether the spread of two populaions are the same or diferent, h0: variance1=variance2, h1: variance1 not equal to variance2.
