Class Notes (1,100,000)
CA (620,000)
McMaster (50,000)
STATS (500)
Lecture 21

STATS 2B03 Lecture 21: confidence intervals and testing for a difference in two means


Department
Statistics
Course Code
STATS 2B03
Professor
Aaron Childs
Lecture
21

This preview shows half of the first page. to view the full 2 pages of the document.
Wearegoingtotakeindependentrandomsamplesfromtwopopulations,andwewanttoestimate
andtestoneof
1.:
2.:
3.:
SmallSampleResults(or)
If
Step1Calculate

̅

̅
1
1
where


iscalledthepooledsamplevariance
Step2
Sameasindexcard,butreplaceby
A100(1)%C.I.foris
̅

̅
,
Assumptions
1.Thesamplesareindependent
2.Bothpopulationsmustbenormal
3.Thepopulationsvariancesmustbeequal()
If
Step1Calculate
̅

̅
Step2Sameasindexcard,butreplacewithwhere
,,

,
,
(replaceby1inCaseI+IIandby1
inCaseIII)
A100(1)%C.I.foris
̅

̅

Assumptions
Sameasabove
Example (continued)
Test::using0.05,whereistheaveragecholesterollevelformen,and
forwomen
̅
215.56,19.14,9
6.4+7.3ConfidenceIntervalsandTestingforaDifferenceinTwo
Means
October2112 7:48PM
Stats 2B03 Pa
g
e 1
You're Reading a Preview

Unlock to view full version