Review on Chapter 4 Part
I
In chapter 4 Part I (p.1-p.25), we have learnt:
1. Sampling distribution of the mean
-Given a population S represented by a random
variable X with mean μ and variance σ 2.
-Consider ALL samples of size n obtained by
selection with replacement (OR without
replacement when S islarge ).
-The sampling distribution of the mean is the
data set
S x =collection of ALLsample means
-It is represented by the random variable X . And
we have
σ2 σ
μ x =μ σ 2x = σ =
n x √n
2. Central Limit Theorem
( n)
2
-If X N ( μ , σ 2 ), then X N μ , σ .
-If X is NOT normal, then by Central Limit
( n ) approximately.
2
Theorem, if n ≥ 30, then X N μ , σ
3. Sampling distribution of the difference between
two means
-Given two populations S1 and S2 represented by
random variables X 1 and X 2 respectively.
-Perform sampling independently on them with
sample sizes n1 and n2 respectively.
-Assume either X 1 and X 2 are both normal, OR the
sample sizes n1 , n2 ≥ 30. (Central Limit Theorem can
be applied!!!!!)
-The sampling distribution of the difference
between two means is the data set
S x − x ={x 1−x 2 : x 1 ∈ S x ∧x 2 ∈ S x }
1 2 1 2
-It is represented by the random variable X 1 −X 2.
And we have
( )
2 2
σ 1 σ2
X 1 −X 2 N μ1−μ2 , +
n1 n 2
P.33
Example 5
Confidence Interval for the Mean (case 1: when σ is
known)
Below are the data in a random sample drawn from
a large population. Construct the 95% confidence
interval for the mean assuming that the population
standard deviation is about 75.
Random Sample
235 374 309 499 253
421 361 514 462 369
394 439 348 344 330
261 374 302 466 535
386 316 296 332 334
Solution:
Step 1:
Sample size n=25
Standard deviation σ =75 (known case)
95% confidence interval α =0.05 Z α / 2=1.96
Step 2: Sample mean
235+374+ …+332+334
x= =370.16
25
Step 3: Margin of error
σ 75
Zα/ 2× =1.96 × =29.4
√n √ 25
Step 4: 95% confidence interval
C . I .=( 370.16−29.4 , 370.16+29.4 )=(340.76 ,399.56)
Interpretation of the above C.I.:
We are 95% confident that the C.I. contains μ.
P.41
Example 6
(a) t 0.10 ,15
(b) t 0.025 ,83
(c) t 0.05 ,196
(d) t 0.005 ,33
(e) t 0.025 ,∞
Solution:
(a) t 0.10 ,15
t 0.10 ,15=1.3406
(b) t 0.025 ,83 (NO v=83 in t table, use the nearest: v=85)
t 0.025 ,83 ≅ t 0.025 , 85=1.9883
(b) t 0.05 ,196 (NO v=196 in t table, use the nearest: v=200)
t 0.05 ,196 ≅ t 0.05 , 200=1.6525
(d) t 0.005 ,33 (NO v=33 in t table, use the nearest: v=35)
t 0.005 ,33 ≅ t 0.005 ,35=2.7238
(e) t 0.025 ,∞
t 0.025 ,∞=1.9600
P.44
Example 7
Confidence Interval for the Mean (case 2: when σ is
unknown)
A random sample of 209 taxpayers was selected in
the US. The mean and S.D. of the income taxes for
the selected individuals are found to be
x=$ 6,001 s=$ 2,864
Estimate the mean income tax amount of all US
taxpayers with 95% confidence.
Solution:
Sample size n=209 Degree of freedom v=n−1=208
Significance level α =1−0.95=0.05
Since NO v=208 in t table, use the nearest one: v=200)
t α / 2 ,v =t 0.025,208 ≅ t 0.025,200
t 0.025,208 ≅ t 0.025,200 =1.9719
Then, calculate the Margin of error:
s 2864
t α / 2 ,v × =1.9719× ≅ 390.6472
√n √209
95% C.I. for the mean income tax amount of all US
taxpayers:
¿ ( 6001−390.6472,6001+390.6472 )
¿(5610.353,6391.647)
P.50
Example 8
(a) χ 20.10 , 5
(b) χ 20.01,100
(c) χ 20.95 , 18
Solution:
(a) χ 20.10 , 5
2
χ 0.10 , 5=9.2364
(b) χ 20.01,100
2
χ 0.01,100 =135.8067
(c) χ 20.95 , 18
2
χ 0.95 , 18=9.3905
P.52
Example 9
Confidence Interval for the Variance
A new machine is installed in a soda bottling plant
for filling 1-liter (=1000 cubic centimeters) bottles.
It is supposed to be so consistent that the variance
of the fills is very small.
A random sample of 25 1-liter fills was taken with
s =0.6333. Estimate the variance of the fills with a 99%
2
confidence.
Solution:
Sample size n=25 Degree of freedom v=n−1=24
Significance level α =1−0.99=0.01
2 2 2 2
χ α /2 , 24= χ 0.005 ,24 χ 1−α /2 ,24 = χ 0.995 ,24
2 2
χ 0.005 , 24=45.5585 χ 0.995 , 24=9.8862
The 99% C.I. for σ 2 :
¿
( ( n−1 ) s 2 ( n−1 ) s 2
,
χ 20.005 ,24 χ 20.995 ,24 )
¿ ( 2445.5585 9.8862 )
( 0.6333 ) 24 ( 0.6333 )
,
¿(0.3336 , 1.5374)