You are on page 1of 15

Ch8 Inference concerning variance

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

Mean Proportion Variance

Z Test t Test Z Test 2 Test


(1 & 2 (1 & 2 (1 & 2 (1 & 2
tail) tail) tail) tail)
8.1 The Estimation of Variances
Standard deviation S
Variance
n

i
( X X ) 2

Let S2 i 1
be the sample variance
n 1

based on any population having variance 2


Unbiased estimation of a population variance

The sample variance


n

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

100(1-a)% confidence interval for


(n 1) S 2 ( n 1) S 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

Alternative Reject null


hypothesis hypothesis if

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

The lapping process which is used to grind certain silicon


wafers to the proper thickness is acceptable only if , the
population standard deviation of the thickness of dice cut
from the wafers, is at most 0.5 mil. Use the 0.05 level of
significance to test the null hypothesis 0.5 against the
alternative hypothesis 0.5 ,if the thickness of 15 dice
cut from such wafers have a standard deviation of 0.64 mil.
Solution
1. Null hypothesis: 0.5

Alternative hypothesis 0.5

2. Level of significance: 0.05

3. Criterion: Reject the null hypothesis if 2


23.685

4. Calculation:
( n 1) S 2
(15 1)(0.64) 2

2
22.94
0 2
0.5 2

5. The null hypothesis cannot be rejected at level 0.05.

6. P-value: 1-0.9387=0.0613 > level of significance


8.3 Hypothesis concerning two variances
If independent random samples of size n1
n2 and are taken from normal
population having the same variance, it
follows from Theorem 6.5 that
S12
F 2
S2
is a random variable having the F
distribution with n1 1 and n2 1 degrees of
freedom
Testing two variances
Null hypothesis 12 22
Criterion Region for testing (Normal population) 2 02

Alternative Test Reject null hypothesis if


hypothesis statistic

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

2. Level of significance: 0.05

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)

5. The null hypothesis must be rejected at level 0.05.

6. P-value: 1-0.965=0.035 < level of significance

You might also like