Chapter-7

Estimation

By Prof. Ashok Bantwa

 Estimation
• Everyone makes estimates
– Estimated marks
– Estimated demand
– Estimated interest rates
– Estimated rainfall etc.
• These estimates are always not perfect
• On the basis of these estimates key decisions are
made
• Therefore effectiveness of decisions depends on the
accuracy of estimates.
 What is estimation ?

• Estimation is related with making inferences about

characteristics of populations from information
contained in sample.
• For example estimating about population mean(  ) by
using sample mean (x )
 Types of estimates

• Two types of estimates can be made about

population,
– Point estimate
– Interval estimate
• Point estimate-
– A point estimate is single number that is used to estimate
an unknown population parameters.
– For. E.g. at this year number of enrollment will be 350
students
 Types of estimates

• Interval estimate-
– An interval estimate is a range of values used to estimate
the population parameter
– For. E.g. at this year number of enrollment will be
between 330 to 380 students
 Estimator and Estimate

• Estimator:
– An estimator is the sample statistic used to estimate
population parameter.
– For e.g. sample mean (x ) can be an estimator of
population mean (  )
 Estimator and Estimate

• Estimate
– Specific numerical value of estimate is known as a
estimate.
 Criteria of Good Estimator
• Unbiasedness

• Efficiency

• Consistency

• Sufficiency
 Point estimate
• Sample mean can be used to estimate population
mean

• Sample standard deviation can be used to estimate

population standard deviation

• Sample proportion can be used to estimate

population proportion
 Interval estimate
• Sample of 200 batteries.
• Sample mean life (x ) is 36 months.
• Now as per point estimate population mean (  ) will
also be 36
• But there is uncertainty associated with it.
• for this we need to find standard error
 
x 
n
10

200
 0.707
It means actual mean will lie in the following
interval
35-0.7017=35.293
35+0.707=36.707
Interval- (35.293,36.707)
• Now we need to find the probability that actual life
will lie in this interval or other interval.
• For this we need to find central limit theorem, as per
this theorem
 Interval Estimate of Proportion From
Large Sample
• Binomial Distribution:
– Experiment involves n identical trials
– Each trial has exactly two possible outcomes: success
and failure
– Each trial is independent of the previous trials
p is the probability of a success on any one trial
q = (1-p) is the probability of a failure on any one trial

  np   npq
Interval Estimate of Proportion From
Large Sample
p  Population Proportion in favor
q  Population Proportion not in favor
p  Sample Proportion in favor
q  Sample Proportion not in favor
 p  Mean of Sampling Distribution of proportion
 p  Standard error of proportion

 p  Estimated Standard error of proportion
Interval Estimate of Proportion From
Large Sample

pq
p  p p 
n

 pq
p
n
 The z Distribution

• Z- distribution is useful only when a distribution is

normal distribution
• We can take a distribution as a normal only when
sample size is more than 30
• If sample size is less than 30 we cant consider it as a
normal distribution.
• In such situation we need to use t-distributiob.
 The t Distribution

• Two conditions precedents for using t- distribution.

– When sample size is 30 or less.

– When population standard deviation is unknown.

• Developed by British statistician, William Gosset
• A family of distributions -- a unique distribution for
each value of its parameter, degrees of freedom (d.f.)
 Characteristics of t- distribution
• More flatter than normal distribution

• Lower at the mean and higher at tails than normal

distribution.
• More values in tails as compared to normal distribution

• As the size of n increases t distribution approximates

normal distribution
Comparison of Selected t Distributions
to the Standard Normal
Standard Normal
t (d.f. = 25)
t (d.f. = 5)
t (d.f. = 1)

-3 -2 -1 0 1 2 3
 Degree of freedom

• Number of values we can choose freely.

• When there are two elements in a sample and we
know the sample mean of these two elements we are
free to specify only one of the element because
other elements will be decided freely.
• Formula for degree of freedom is (n-1)
 Using t-distribution table
• T-table is more compact and shows area and t
values for only few percentages (10,5,2 and 1
percentage)
• It does not focus on the chance that the population
parameter being estimated will fall within our
confidence interval, instead it measures the chance
that population parameter will not be within our
interval
• We must specify the degree of freedom with which
we are dealing
 The t Distribution
• t formula

x
t
x
 Table of Critical Values of t

df t0.100 t0.050 t0.025 t0.010 t0.005

1 3.078 6.314 12.706 31.821 63.656
2 1.886 2.920 4.303 6.965 9.925
3 1.638 2.353 3.182 4.541 5.841
4 1.533 2.132 2.776 3.747 4.604 
5 1.476 2.015 2.571 3.365 4.032

23 1.319 1.714 2.069 2.500 2.807

24 1.318 1.711 2.064 2.492 2.797
25 1.316 1.708 2.060 2.485 2.787
 t
29 1.311 1.699 2.045 2.462 2.756
30 1.310 1.697 2.042 2.457 2.750

40 1.303 1.684 2.021 2.423 2.704 With df = 24 and = 0.05,

60
120
1.296
1.289
1.671
1.658
2.000
1.980
2.390
2.358
2.660
2.617
t = 1.711.
 1.282 1.645 1.960 2.327 2.576
Confidence Intervals for  of a Normal
Population: Unknown 
s
xt
n
or
s s
x t    xt
n n
df  n  1
 Example 3
The owner of a large equipment rental company wants to
make a rather quick estimate of the average number of days
a piece of ditchdigging equipment is rented out per person
per time. The company has records of all rentals, but the
amount of time required to conduct an audit of all accounts
would be prohibitive. The owner decides to take a random
sample of retail invoices. Fourteen different rentals of
ditchdiggers are selected randomly from the files, yielding
the following data. She uses these data to construct a 99%
confidence interval to estimate the average number of days
that a ditchdigger is rented and assumes that the number of
days per rental is normally distributed in the population.
3 1 3 2 5 1 2 1 4 2 1 3 1 1
 Solution to Example 3
x  2.14, s  1.29, n  14, df  n  1  13
 1  .99
  0.005
2 2
t .005,13  3.012
s s
xt    xt
n n
1.29 1.29
2.14  3.012    2.14  3.012
14 14
2.14  1.04    2.14  1.04
1.10    3.18
 Solution to Example 3
s s
x t    xt
n n
1.29 1.29
2.14  3.012    2.14  3.012
14 14
2.14  1.04    2.14  1.04
1.10    3.18
Confidence Interval to Estimate
the Population Proportion
pˆ qˆ pˆ qˆ
pˆ  z   p  pˆ  z 
2 n 2 n
where :
pˆ = sample proportion
qˆ = 1 - pˆ
p = population proportion
n = sample size
 Example 4
A clothing company produces men’s jeans. The
jeans are made and sold with either a regular cut or
a boot cut. In an effort to estimate the proportion of
their men’s jeans market in Oklahoma City that
prefers boot-cut jeans, the analyst takes a random
sample of 212 jeans sales from the company’s two
Oklahoma City retail outlets. Only 34 of the sales
were for boot-cut jeans. Construct a 90% confidence
interval to estimate the proportion of population in
Oklahoma City who prefer boot-cut jeans.
 Solution to Example 4
x 34
n  212, x  34, pˆ    0.16
n 212
qˆ = 1 - pˆ  1  0.16  0.84
90% Confidence  z  1.645

pˆ qˆ pˆ qˆ
pˆ  z  p  pˆ  z
n n
(0.16)(0.84) (0.16)(0.84)
0.16  1.645  p  0.16  1.645
212 212
0.16  0.04  p  0.16  0.04
0.12  p  0.20
 Determining Sample Size
when Estimating 
• z formula x
z

n

• Error of Estimation (tolerable

E  x
error)
2
• Estimated Sample Size
z 2   z 2  
2 2

n  
E
2
 E 
 
• Estimated  1
 range
4
Sample Size When Estimating  Example
E  1,   4
90% confidence  z  1.645

z 
2 2

n 2
2
E
(1.645) 2 (4) 2

12
 43.30 or 44
 Example 5
Suppose you want to estimate average age of all
Boeing 727 airplanes now in active domestic U.S.
service. You want to be 95% confident, and you
want your estimate to be within two years of the
actual figure. The 727 was first placed in service
about 30 years ago, but you believe that no active
727s in the U.S. domestic fleet are more than 25
years old. How large a sample should you take?
 Solution to Example 5
E  2, range  25
95% confidence  z  1.96
1 1
estimated  : range    25  6.25
4  4

z
2 2

n 2
E
(1.96) 2 (6.25) 2

22
 37.52 or 38
 Determining Sample Size
when Estimating p
• z formula pˆ  p
Z
pq
• Error of Estimation (tolerable
n error)

• Estimated Sample Size

E  pˆ  p

z 2 pq
n 2
E
 Example 6
Hewitt Associates conducted a national survey to determine
the extent to which employers are promoting health and
fitness among their employees. One of the questions asked
was, Does your company offer on-site exercise classes?
Suppose it was estimated before the study that no more that
40% of the companies would answer Yes. How large a
sample would Hewitt Associates have to take in estimating
the population proportion to ensure a 98% confidence in the
results and to be within .03 of the true population
proportion?
Solution to Example 6
E  0.03
98% Confidence  Z  2.33
estimated P  0.40
Q  1  P  0.60

z 2 pq
n
E2
(2.33) 2 (0.40)(0.60)

(.03) 2
 1,447.7 or 1,448
 Determining Sample Size when
Estimating p with No Prior Information
p pq 400 z = 1.96
350 E = 0.05
0.5 0.25
300
0.4 0.24 250
n 200
0.3 0.21
150
0.2 0.16 100
50
0.1 0.09
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
P
2 1
z 4
n 2
E
 Example: Determining n when
Estimating p with No Prior Information
E  0.05
90% Confidence  z  1.645
with no prior estimate of p, use p  0.50
q  1  p  0.50

n z pq
2
E
(1.645) 2 (0.50)(0.50)

(.05) 2
 270.6 or 271