BS 6-8

Business Statistics
Interval Estimation
Prashanth V. Anand
PhD, IIM Lucknow
Assistant Professor, IIM Amritsar
Point and Interval Estimates..
2
Point and Interval Estimates..
• How much uncertainty is associated with a point estimate of a

population parameter?
• An interval estimate provides more information about a population

characteristic than does a point estimate.
• Such interval estimates are called confidence intervals. 3

Confidence Interval Estimate..
• An interval gives a range of values:
• Takes into consideration variation in sample statistics from
sample to sample.
• Based on observations from 1 sample.
• Gives information about closeness to unknown population
parameters.
• Stated in terms of level of confidence:
• e.g. 95% confident, 99% confident.
• Can never be 100% confident.
4
Dessert Fill Video example
• Population has µ = 368 grams and σ = 15 grams.
• If you take a sample of size n = 25, you know:

• Sample Mean (X-bar) = 362.3
• Population σ = 15 grams.
• When you don’t know µ, you use X-bar to estimate µ based on CI (95% assumed).
• If X = 362.3 the interval is 362.3 ± 1.96 * 15 / 5 = (356.42, 368.18). (1.96 is the Z value
for 95% CI – we will see how to calculate this later)
• Since 356.42 ≤ µ ≤ 368.18 the interval based on this sample makes a correct statement
about µ.
But what about the intervals from other possible samples of size 25?
5
Lower Upper Contain
Sample # X
Limit Limit µ?
1 362.30 356.42 368.18 Yes
2 369.50 363.62 375.38 Yes
3 360.00 354.12 365.88 No
4 362.12 356.24 368.00 Yes
5 373.88 368.00 379.76 Yes
6
• In practice you only take one sample of size n.

• In practice you do not know µ so you do not know if the interval actually
contains µ.
• However you do know that 95% of the intervals formed in this manner will
contain µ.
• Thus, based on the one sample you actually selected, you can be 95% confident
your interval will contain µ (this is a 95% confidence interval).
7
Estimation Process..
8
Estimation Process..
9
Confidence Level..
• Confidence interval will contain the unknown population parameter.
• A percentage (less than 100%).
• Suppose confidence level = 95%.
• Also written (1 - ) = 0.95, (so  = 0.05).
• A relative frequency interpretation:
• 95% of all the confidence intervals that can be constructed will contain the
unknown true parameter.
• A specific interval either will contain or will not contain the true
parameter:
• No probability involved in a specific interval.
10
Confidence Interval.. Critical Value
11
Confidence Interval.. μ (σ Known)
12
13
14
Confidence Level.. μ (σ Known)
15
16
17
Confidence Level.. μ (σ Unknown)
18
Confidence Interval.. μ (σ Unknown)
Can we really know σ?
Probably not!
In virtually all real world business situations, σ is not known.
If there is a situation where σ is known then µ is also known

(since to calculate σ you need to know µ.)
If you truly know µ there would be no need to gather a sample

to estimate it.
19
• If the population standard deviation σ is unknown, we can

substitute the sample standard deviation, S.
• This introduces extra uncertainty, since S is variable from

sample to sample.
• So we use the t distribution instead of the normal

distribution.
20
21
Student’s t distribution
22
23
24
25
26
27
Confidence Interval..
28
Confidence Interval.. π
• An interval estimate for the population proportion ( π ) can be
calculated by adding an allowance for uncertainty to the sample
proportion (p).
• The distribution of the sample proportion is approximately normal if
the sample size is large, with standard deviation:
• We will estimate this with sample data:
29
30
31
32
33
Determining Sample Size..
34
35
36
37
38
39
40
41
42
How large a sample would be necessary to

estimate the true proportion of sales invoices
containing errors in a large population within
±7%, with 95% confidence?
(Assume a pilot sample yields p = 0.15.)
43
44
Confidence Interval.. Finite Population
45
Confidence Interval.. Finite Population
Suppose N = 1000, n = 100, X = 50, s = 10
s N −n
95% CI for  : X  t / 2 =
n N −1
10 1000 − 100
50  1.984 =
100 1000 − 1
50  1.88 = (48.12, 51.88)
46
Sample Size.. Finite Population
47
Confidence Interval..
48
Confidence Interval.. Application
• Six advantages of statistical sampling in auditing:
• Sampling is less time consuming and less costly.
• Sampling provides an objective way to calculate the sample size in advance.
• Sampling provides results that are objective and defensible.
• Because the sample size is based on demonstrable statistical principles, the audit is
defensible before one’s superiors and in a court of law.
• Sampling provides an estimate of the sampling error:
• Allows auditors to generalize their findings to the population with a known sampling error.
• Can provide more accurate conclusions about the population.
• Sampling is often more accurate for drawing conclusions about large populations:
• Examining every item in a large population is subject to significant non-sampling error.
• Sampling allows auditors to combine, and then evaluate collectively, samples
collected by different individuals.
49
Confidence Interval.. Population Total Amount
50
Confidence Interval.. Population Total Amount
A firm has a population of 1,000 accounts and wishes to estimate the total
population value.
A sample of 80 accounts is selected with average balance of $87.6 and

standard deviation of $22.3.
Find the 95% confidence interval estimate of the total balance.

S N−n
NX  N (t  / 2 )
N = 1,000 , n = 80, X = 87 .6, S = 22.3 n N −1
22.3 1,000 − 80
= (1,000)(87.6)  (1,000)(1.9905)
80 1,000 − 1
= 87,600  4,762.48
The 95% confidence interval for the population total
51
balance is $82,837.52 to $92,362.48.
Confidence Interval.. Total Difference
52
Confidence Interval.. Total Difference
53
Confidence Interval.. One Sided
54
Confidence Interval.. Bootstrapping
Bootstrapping Is A Method To Use When Population Is Not Normal
To estimate a population mean using bootstrapping, you would:

1. Select a random sample of size n without replacement from a population
of size N.
2. Resample the initial sample by selecting n values with replacement from
the initial sample.
3. Compute X from this resample.
4. Repeat steps 2 & 3 m different times.
5. Construct the resampling distribution of X.
6. Construct an ordered array of the entire set of resampled X’s.
7. In this ordered array find the value that cuts off the smallest α/2(100%)
and the value that cuts off the largest α/2(100%). These values provide the
lower and upper limits of the bootstrap confidence interval estimate of μ.
55
56
Example: Processing Time of Insurance Applications
57
Example: Processing Time of Insurance Applications
58
59
60

BS 6-8

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

BS 6-8

Uploaded by

Copyright:

Available Formats

Business Statistics

• How much uncertainty is associated with a point estimate of a

• An interval estimate provides more information about a population

• Such interval estimates are called confidence intervals. 3

• If you take a sample of size n = 25, you know:

2 369.50 363.62 375.38 Yes

3 360.00 354.12 365.88 No

4 362.12 356.24 368.00 Yes

5 373.88 368.00 379.76 Yes

• In practice you only take one sample of size n.

In virtually all real world business situations, σ is not known.

If there is a situation where σ is known then µ is also known

If you truly know µ there would be no need to gather a sample

• If the population standard deviation σ is unknown, we can

• This introduces extra uncertainty, since S is variable from

• So we use the t distribution instead of the normal

• We will estimate this with sample data:

How large a sample would be necessary to

A sample of 80 accounts is selected with average balance of $87.6 and

Find the 95% confidence interval estimate of the total balance.

To estimate a population mean using bootstrapping, you would:

You might also like