STAT 263 Lecture Notes - Lecture 8: Statistical Hypothesis Testing, Type I And Type Ii Errors, Null Hypothesis
4 Dec 2017
School
Department
Course
Professor
Document Summary
Ha: /= 0 one- tailed test one- tailed test two- tailed test http://www. mathnstuff. com/math/spoken/here/2class/90/htest. htm. Reduction and absurdum: if a is true then b must be true. Thus if it is known that b is not true then a must not be true. In a hypothesis test: we would say a is probably not true (used in stats) Means of large samples, theoretical formula: Level of significance: prob of making a type 1 error which is acceptable to the experimenter. Det before examining the data. Academic default setting is 5% but is not appropriate for real life. Ex. the mean of a pop is commonly believed to be 45. 6. Is there evidence at the. P(z>0. 86)=0. 1949 is the p- value, this value does not fall in the rejection of null hypothesis. Anything higher than 1. 645 could be rejected. We don"t have evidence at the 5% level to conclude the pop mean is greater than 45. 6.
