STATS 2B03 Study Guide - Quiz Guide: Type I And Type Ii Errors, Statistical Hypothesis Testing

• School: McMaster University
• Department: Statistics
• Course: STATS 2B03
• Professor: Aaron Childs

Document Summary

Can never conclude ha is definitely true. Doesn"t mean you accept h0 (or that it is probably true) Conclude that there is not sufficient evidence to reject h0, or that there is not enough strong evidence in favour of ha. Analogy: h0 no guilty (not strong enough evidence for guilt), ha: guilty. The truth h0 is true ha is true. Example: suppose 15 students are sampled and find that xavg=78. 6 and reject. H0 72 but the test average ends up being 63. What error is that? (or is it a correct decision?) But for a given significance level, increasing the sample size always decreases , and increases the power. Procedure for testing about a population mean ( ) for n 30 step 1: calculate. Replace ( ) with (s) if ( ) is unknown step 2: Case i: ha: > 0 reject h0 if z>z1- . Case ii: ha: < 0 reject h0 if z