STAT 263 Lecture Notes - Lecture 7: Type I And Type Ii Errors, Alternative Hypothesis, Statistical Hypothesis Testing
4 Dec 2017
School
Department
Course
Professor
Document Summary
Statistical hypothesis: assertion or conjecture about a parameter of any pop, may concern type or nature of a pop. Hypothesis of no change, difference or relationship. Alternative hypothesis (ha): referred to as the research hypothesis, formulated together w null, it is accepted if null is rejected. X = x = 0. 08/ 40 =0. 0126. If h0 is false= type 2 error w prob b. If ha is false= type 1 error w prob . 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) This is the best we can do since no other info is given. 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.
