STAT 135 Study Guide - Final Guide: Jerzy Neyman, Statistical Hypothesis Testing, Itta Of Metz
Final Review 4130
Hypothesis testing
9.2 Neyman Pearson
accept lreject
Fisher 7- value
Bayesian Putolx )vs Patil ×)
outline :1) null a) "
results to chance "
b) specify nun dist (at least exp value us )
)
Coften In terms of parameters )
simple :specify value M=o
2) alternative
WMonte :range of values u=o
a) "not due to chance "
b) Most commonly composite (n>o)
determine one tailed vs two tailed (uto )
37 test statistics
41 p-value is chance of data like we got ,
Or more extreme ,given null is true ,the smaller
the P-value ,the stronger the evidence against null
5) consider a) reject null ?(what is L)
b) Coin results be explained by chance ?
Could be chance ,doesn't look like Chance
not due to chance
c) sth particular about the problem
9.1 LR Test PCHoK)_ CHo ,H,simple )
T.CH .lx )
