Confidence Interval

Estimation
 Statistical Methods

Statistical
Methods

Descriptive Inferential
Statistics Statistics

Hypothesis
Estimation
Testing
 Inference Process

Estimates Population
& Tests

Sample
Statistics
X, Ps Sample
Population Parameters Are Estimated
with Point Estimator
Estimate Population with Sample
Parameter... Statistic
Mean  X
Proportion p ps
2
Variance 2 s
Differences 12 X1  X 2
 Confidence Interval Estimation
Process
I am 95%
Population Random Sample confident
that  is
Mean
Mean, , is between
unknown X = 50
40 & 60.
 Confidence Interval Estimates
Provide range of values
– Takes into consideration variation in
sample statistics from sample to sample
– Is based on observation from one sample
– Gives information about closeness to
unknown population parameters
– Is stated in terms of level of confidence
Never 100% certain
 Elements of Confidence
Interval Estimation
A Probability That the Population Parameter
Falls Somewhere Within the Interval.
Sample
Confidence Interval Statistic

Confidence Confidence
Limit (Lower) Limit (Upper)
 Confidence Intervals
X  Z  X  X  Z  
n
x_

X
  2.58 X  1.645 X   1.645 X   2.58 X
 1.96 X  1.96 X
90% Samples
95% Samples
99% Samples
 Level of Confidence
• Probability that the unknown population parameter
falls within the interval
• Denoted (1 -  level of confidence
Is Probability That the Parameter Is Not Within
the Interval
• The typical values are 99%, 95%, 90%.
Corresponding Z values are 2.58, 1.96, 1.645 (for
two side problems).
Experiment: throw 2 dice, use mean 2 to construct
a 94.5% confidence interval estimate for .
 Interval and Level of Confidence
Sampling Distribution of the Mean
X
  Z / 2 X  /2   Z / 2 X
1  /2

Intervals X
extend from X  
1   100%
X  Z X of intervals
constructed
to contain  ; 
100% do not.
X  Z X
Confidence Intervals
 Factors Affecting
Interval Width
1.Data Dispersion Intervals Extend from
X  Z  X to X  Z  X
Measured by 
2.Sample Size
X  / n
3.Level of Confidence
(1 - )
Affects Z
© 1984-1994 T/Maker Co.
 Confidence Interval Estimates

Confidence
Intervals

Mean Proportion

 Known  Unknown
 Confidence Interval Estimate
For Mean ( Known)
Assumptions
– Population standard deviation () is known
– Population is normally distributed
– If not normal, can be approximated by
normal distribution (n  30)
Confidence Interval Estimate
 
X  Z / 2     X  Z / 2 
n n
 Estimation Example
For Mean ( Known)
Population is normally distributed with  = 10.
The mean of a random sample of n = 25 is X =
50. Set up a 95% confidence interval estimate
for .
X  Z      X  Z  
2 n 2 n

501.96 10    501.96 10
25 25
46.08    53.92
 Confidence Interval Estimate
Mean ( Unknown)
Assumptions
Population Standard Deviation () Is
Unknown
Population Must Be Normally Distributed
Use Student’s t Distribution
Confidence Interval Estimate:
S S
X  t  / 2, n 1     X  t  / 2, n 1 
n n
 Student’s t Distribution
Degree of freedom
Standard
Normal = n 1
Bell-Shaped
t (df = 13)
Symmetric
‘Fatter’ Tails t (df = 5)

Z
0 t
 Student’s t Table
Assume: n = 3
/2 df = n - 1 = 2
Upper Tail Area  = .10
df .2 .10 /2 =.05
.05
5
1 1.000 3.078 6.314

2 0.817 1.886 2.920 /2 =.05

3 0.765 1.638 2.353

0 2.920 t
t Values
 Degrees of Freedom (df)
1. Number of Observations that Are Free to Vary
After Sample Statistic Has Been Calculated
2. Example
Sum of 3 Numbers Is 6
X1 = 1 (or Any Number)
X2 = 2 (or Any Number)
X3 = 3 (Cannot Vary)
Sum = 6
 Estimation Example
Mean ( Unknown)
Population is normally distributed . A random
sample of n = 25 has X = 50 and S = 8. Set
up a 95% confidence interval estimate for .
S S
X  t  / 2, n 1     X  t  / 2, n 1 
n n
8 8
50  2.0639     50  2.0639 
25 25
46.69    53.30
 Confidence Interval Estimate
for Proportion
Assumptions
Two categorical outcomes
Population follows binomial distribution
Normal approximation can be used if
n·p  5 and n·(1 - p)  5
Confidence interval estimate:
pS 1  pS  pS 1  pS 
pS  Z / 2  p  pS  Z / 2
n n
 Estimation Example: Proportion
A random sample of 400 voters showed 32
preferred candidate A. Set up a 95% confidence
interval estimate for p.
Ps = 32/400 = .08
ps  (1  ps ) ps  (1  ps )
ps  Z  / 2   p  ps  Z  / 2 
n n
.08  (1.08 ) .08  (1.08)
.08  196
.   p  .08  196
. 
400 400
.053  p  .107
 Finding Sample Sizes for the Mean

Z  X-   Error

X X I don’t want to
Error  Z   Z   sample too much
X n or too little!

Z 2  2
n
Error 2
Require:
1) the confidence level desired
when 1-  = .95, Z = 1.96
2) the sampling error permitted
3) the standard deviation, 
 Sample Size for the Mean
Suppose you worked in Human Resources
and planned to survey employees to find their
average medical expenses. You wanted to be
95% confident that the sample mean is
within ±$50. A pilot study showed that 
was $400. What sample size do you use?
2 2 2 2
Z  (1.96) (400)
n 2
 2
 245.86  246
Error (50)
Always round up
 Sample Size for Proportion
• Define the sampling error, Error = ps - p, then the
Sample Size for a Proportion can be determined by
the following equation:
2
Z p (1  p )
n 2
Error
• the value of Z and Error are given
• the value of p can be determined according to
historical information
• Assume p = 0.5 conservatively if no historical data
are available
 Determining Sample Size for
Proportion
Out of a population of manufacturing products,
30% are defective. If we randomly select a
sample, what sample size is needed to be within
± 5% with 90% confidence?
Z p 1  p  1.645  0.3  0.7 
2 2

n 2
 2
Error 0.05
 227.3  228
Round Up