Professional Documents
Culture Documents
A First Course
(2nd Edition)
Chapter 5
Sampling Distributions and
Confidence Interval Estimation
2000 Prentice-Hall, Inc.
Chap. 5 - 1
Developing
Sampling Distributions
Suppose theres a
population...
Population size, N = 4
Random variable: X
is Age of individuals
Values of X: 18, 20, 22,
24 measured in years
Chap. 5 - 2
Developing
Sampling Distributions
(Continued)
Xi
P(X)
.3
i 1
18 20 22 24
21
4
.2
.1
0
Xi
i 1
2.236
(18)
(20)
(22)
(24)
Uniform Distribution
Chap. 5 - 3
Developing
Sampling Distributions
(Continued)
2nd Observation
18
20
22
24
16 Sample Means
1st 2nd Observation
Obs 18 20 22 24
18 18 19 20 21
20 19 20 21 22
16 Samples Taken
with Replacement
22 20 21 22 23
24 21 22 23 24
Chap. 5 - 4
Developing
Sampling Distributions
(Continued)
18 18 19 20 21
20 19 20 21 22
P(X)
.3
.2
22 20 21 22 23
.1
24 21 22 23 24
Sample size = 2
2000 Prentice-Hall, Inc.
Sample Means
Distribution
18 19
20 21 22 23
24
# of samples = 16
Chap. 5 - 5
Developing
Sampling Distributions
(Continued)
Xi
18 19 19 24
x
21
N
16
i 1
X i x
N
i 1
18 21 2 19 21 2 24 21 2
16
1.58
Chap. 5 - 6
= 21
= 2.236
x 21
P(X)
.3
P(X)
.3
.2
.2
.1
.1
(18)
(20)
(22)
D X
18 19
x 1.58
20 21 22 23
24
(24)
Chap. 5 - 7
X
n
As n increases X decreases
Chap. 5 - 8
Population Distribution
= 10
= 50
Variation
X
n
Sampling with
Replacement
2000 Prentice-Hall, Inc.
Sampling Distributions
n=4
X = 5
n =16
X = 2.5
X-X = 50
X
Chap. 5 - 9
Sampling
Distribution of X
Becomes
Almost Normal
regardless of
shape of
population
XX
2000 Prentice-Hall, Inc.
Chap. 5 - 10
Population Distribution
= 10
X
Variation
X
n
Sampling with
Replacement
2000 Prentice-Hall, Inc.
= 50
X
Sampling Distributions
n=4
X = 5
n =30
X = 1.8
X 50
X
Chap. 5 - 11
Estimation Process
Population
Mean, , is
unknown
Random Sample
Mean
X = 50
I am 95%
confident that
is between 40 &
60.
Sample
Chap. 5 - 12
Population Parameters
Estimated
Estimate Population
Parameter...
Mean
Proportion
Variance
Difference
2000 Prentice-Hall, Inc.
with Sample
Statistic
_
X
-
1
ps
s
2
_
_
x - x
1
2
Chap. 5 - 13
Chap. 5 - 14
Elements of Confidence
Interval Estimation
A Relative Frequency That the Population Parameter
Falls Somewhere Within All Intervals constructed.
Sample
Range (Closeness)
Statistic
Confidence Limit
(Lower)
Confidence Limit
(Upper)
5% Uncertainty
Chap. 5 - 15
X Error
X = Error = X
Error
Error Z x
X Z X
1984-1994 T/Maker Co.
2000 Prentice-Hall, Inc.
Chap. 5 - 16
Confidence Intervals
X Z X X Z
n
x_
_
X
1.645 x 1.645 x
90% of the Samples
1.96 x
2.58 x
2000 Prentice-Hall, Inc.
1.96 x
95% of the
Samples
2.58 x
Chap. 5 - 17
Level of Confidence
Relative frequency that the unknown
population parameter falls within
intervals constructed
Denoted (1 - ) % = level of confidence
2000 Prentice-Hall, Inc.
Chap. 5 - 18
Intervals &
Level of Confidence
Sampling
Distribution /2
of the Mean
Intervals
Extend from
1 -
/2
to
(1 - ) % of
Intervals
Constructed
Contain .
% Do Not.
X Z X
X Z X
Confidence Intervals
Chap. 5 - 19
Factors Affecting
Interval Width
Data variation
Measured by
X - Z
to X + Z
Sample size
X
n
Level of confidence
(1 - )
1984-1994 T/Maker Co.
Chap. 5 - 20
Known
Proportion
Unknown
Chap. 5 - 21
Population is normally
distributed
X Z / 2
X Z / 2
n
n
Chap. 5 - 23
Known
Proportion
Unknown
Chap. 5 - 24
Chap. 5 - 25
Chap. 5 - 26
Students t Distribution
Standard
Normal
Bell-Shaped
Symmetric
Fatter Tails
t (df = 13)
t (df = 5)
0
2000 Prentice-Hall, Inc.
Z
t
Chap. 5 - 27
Mean of 3 numbers is 2
X1 = 1 (or any number)
X2 = 2 (or any number)
X3 = 3 (cannot vary)
mean = 2
=2
Chap. 5 - 28
Students t Table
Upper Tail Area
df
.25
.10
.05
Assume: n = 3
df = n - 1 = 2
= .10
/2 =.05
/ 2 = .05
t Values
2000 Prentice-Hall, Inc.
2.920
Chap. 5 - 29
S
X t / 2 ,n1
n
50 2.0639
8
25
46 . 69
S
X t / 2 ,n1
n
50 2.0639
8
25
53 . 30
Chap. 5 - 30
Known
Proportion
Unknown
Chap. 5 - 31
Np 5
&
n(1 - p) 5
Chap. 5 - 32
ps ( 1 ps )
ps Z / 2
n
Chap. 5 - 33
ps 1 ps
n
.08 1 .08
.08 1.96
400
ps Z
ps 1 ps
n
.08 1 .08
p .08 1.96
400
.053 p .107
Chap. 5 - 34
Chap. 5 - 35
Chap. 5 - 36