class10-1-handouts

class10-1-handouts - PSTAT 120B - Probability &...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
Class # 10-1- Hypothesis testing review Jarad Niemi University of California, Santa Barbara 2 June 2010 Jarad Niemi (UCSB) Hypothesis testing review 2 June 2010 1 / 20
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Class overview Announcements Announcements Homework Homework 7 due now Final exam Monday, 7 June @ 8am-11am in HFH 1104 (here) Allowed/required items: one (front and back) 8 1 2 × 11 cheat sheet calculator #2 pencil Jarad Niemi (UCSB) Hypothesis testing review 2 June 2010 2 / 20
Background image of page 2
Class overview Goals P-values Review of hypothesis testing Jarad Niemi (UCSB) Hypothesis testing review 2 June 2010 3 / 20
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
P-values Definition If W is a test statistic, the p-value is the smallest level of significance α for which the observed data indicate that the null hypothesis should be rejected. Definition The p-value is the probability of obtaining a test statistic W at least as extreme as the one observed assuming the null hypothesis is true. Jarad Niemi (UCSB) Hypothesis testing review 2 June 2010 4 / 20
Background image of page 4
Hypothesis testing decisions Any kind of decision you make under uncertainty can have 4 possible outcomes. H 0 True False Test result Reject H 0 Type I error True positive Do not reject H 0 True negative Type II error α = P(type I error) β = P(type II error) 1 - β = power of the test Jarad Niemi (UCSB) Hypothesis testing review 2 June 2010 5 / 20
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Large sample tests Large-sample α -level hypothesis tests H 0 : θ = θ 0 H a : θ > θ 0 (upper-tail alternative) θ < θ 0 (lower-tail alternative) θ 6 = θ 0 (two-tailed alternative) Test statistic : z = ˆ θ - θ 0 σ ˆ θ Rejection region : z > z α (upper-tail RR) z < - z α (lower-tail RR) | z | > z α/ 2 (two-tailed RR) p-values : P ( Z > z ) (upper-tail) P ( Z < z ) (lower-tail) 2 P ( Z > | z | ) (two-tailed) Jarad Niemi (UCSB) Hypothesis testing review 2 June 2010 6 / 20
Background image of page 6
Sample size Typically, the type I error probability α is given, and the type II error probability β can be chosen by selecting the appropriate sample size n . To obtain a specific β find the smallest integer n , such that n ( z α + z β ) 2 σ 2 ( θ 0 - θ a ) 2 Jarad Niemi (UCSB) Hypothesis testing review 2 June 2010 7 / 20
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Sample size Relationship to confidence intervals Type Large-sample CIs Large-sample hypothesis testing Pivotal Interval Rejection region Upper-tail ˆ θ - θ σ ˆ θ < z α ˆ θ - θ 0 σ ˆ θ > z α Lower-tail ˆ θ - θ σ ˆ θ > - z α ˆ θ - θ
Background image of page 8
Image of page 9
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 20

class10-1-handouts - PSTAT 120B - Probability &amp;...

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online