Class Notes (1,100,000)
CA (620,000)
McMaster (50,000)
STATS (500)
Lecture 20

# STATS 2B03 Lecture 20: f-test for variances

Department
Statistics
Course Code
STATS 2B03
Professor
Aaron Childs
Lecture
20

This preview shows half of the first page. to view the full 1 pages of the document.
Wearegoingtotakeindependentrandomsamplesfromtwopopulations,andwewanttotest:
:

:

Procedure
Step1Calculate
,
whereisthelarger variance,andisthecorrespondingsamplesize
Step2Reject
, 
,,
(usetheFTableisTableGonpageA42)
Assumptions:
Bothpopulationsarenormallydistributed
Example:
̅
 215.56,with19.14
̅
 211.18,with32.25
Theabovehypothesistestswhethertheamountofvariationincholesterollevelsisdifferentformen
thanforwomen.
use0.05
isnotnecessarilytheonecalled;it'sthelargerofthetwo values!
, 32.25
19.14
 2.839
Donotrejectsince2.839 4.30 sowecannot concludethatthepopulationvariancesaredifferent
7.8FTestforVariances
October2012 1:00AM
Stats 2B03 Pa
g
e 1

Unlock to view full version