Professional Documents
Culture Documents
Dr. Deshi Ye
yedeshi@zju.edu.cn
Outline
The estimation of variance
Hypothesis concerning one variance
Hypothesis concerning two variances
One Population Tests
One
Population
i
( X X ) 2
Let S2 i 1
be the sample variance
n 1
i
( X X ) 2
S2 i 1
n 1
is an unbiased estimator of 2
Remark: the sample standard deviation S is not
an unbiased estimator of
However, for large samples the bias is small, and it is common
practice to estimate with S
Confidence interval
By Theorem 6.4
(n 1) S 2
2
is a random variable having the chi-square distribution
with n-1 degrees of freedom.
(n 1) S 2
2
1 / 2 / 2
2
/ 2
2
12 / 2
8.2 Hypothesis concerning one variance
Consider the problem of testing the null
hypothesis that a population variance
equals a specified constant against a
suitable one-sided or two-sided
alternative.
H
Null Hypothesis 0 : 2
2
0
( n 1) S 2
is a random sample from a normal population with
2
the variance 0 is a random variable having the chi-
2
02
square distribution with n-1 degree of freedom.
Criterion Region for testing (Normal population) 2 02
2 02 2 12
2 2 2 2
0
2 12 / 2 or 2 2 / 2
2 02
EX. Testing hypothesis concerning a standard deviation
4. Calculation:
( n 1) S 2
(15 1)(0.64) 2
2
22.94
0 2
0.5 2
12 22 S 22
F 2
F F (n2 1, n1 1)
S1
2 2 S12 F F (n1 1, n2 1)
1 2 F 2
S2
S M2
F 2 F F (nM 1, nm 1)
2
1
2
2
Sm
EX.
It is desired to determine whether there is less variability
in the silver plating done by Company 1 than in that done
by Company 2. If independent random samples of size
12 of the two companies work yield s1 0.035 and
s2 0.062 12 22
, test the null hypothesis
12 22 against
the alternative hypothesis at the 0.05 level of
significance.
Solution
1. Null hypothesis: 12 22
Alternative hypothesis 12 22
S 22
3. Criterion: Reject the null hypothesis if F 2.82, F 2
2
S1
(0.062)
4. Calculation: F 2
3.14
(0.035)