Confidence intervals

Chapters 11, 13, 14

Confidence Intervals
Learning Objectives
1)Calculate a confidence interval for a
proportion/mean.
2)Calculate a confidence interval for the
difference between two proportions/mean.
3)Choose the size of a sample to achieve a given
level of accuracy.
 Review: Types of Estimators
Point Estimate
ü A single-valued estimate.
ü A single element chosen from a sampling distribution.
ü Conveys little information about the actual value of the population
parameter, about the accuracy of the estimate.

Confidence Interval or Interval Estimate

ü An interval or range of values believed to include the unknown
population parameter.
ü Associated with the interval is a measure of the confidence we have
that the interval does indeed contain the parameter of interest.
 Estimation
• We use the sample statistics to make inferences about the unknown
population parameters (e.g., the mean / proportion).
• A confidence interval, or interval estimate, provides a range of values that, with
a certain level of confidence, contains the population parameter of interest.
• To construct a confidence interval for the mean/proportion, it is essential that
the sampling distributions of 𝑋! and 𝑃! follow (approximately) a normal
distribution. (CLT)
• A confidence interval is associated with a margin of error that accounts for the
standard error of the estimator and the desired confidence level of the interval.
• The confidence interval for the population mean/proportion is constructed as:
point estimate ± margin of error.
Confidence interval for mean when 𝜎 is known.
The idea is that we want an upper (𝑈) and lower (𝐿)
boundary as of an interval expressed as a function of the
# such that
point estimate X

𝑃 𝐿 <𝜇 <𝑈 =1 −𝛼
 Confidence interval for mean when 𝜎 is known.
If the sample is sufficiently large (≥30),
regardless of the shape of the
population distribution, the sampling
distribution is normal (Central Limit
Theorem). We use the z distribution.

𝑃 −𝑧! < 𝑍 < 𝑧! = 1 − 𝛼

" "
Confidence interval for mean when 𝜎 is known.
𝑋% − 𝜇
𝑍= 𝜎
𝑛

𝑋/ − 𝜇
𝑃 −𝑧! < 𝜎 < 𝑧! = 1 − 𝛼
" "
𝑛

𝜎 𝜎
𝑃 𝑋/ − 𝑧! × < 𝜇 < 𝑋/ + 𝑧! × =1 −𝛼
" 𝑛 " 𝑛
Confidence interval for mean when 𝜎 is known.
A 100 1 − 𝛼 % confidence interval for the population
mean 𝜇 when 𝜎 is known is computed as the below.
𝜎
𝑥̅ ± 𝑧"⁄#
𝑛
𝜎 𝜎
𝑋% − 𝑧" × < 𝜇 < 𝑋% + 𝑧" ×
# 𝑛 # 𝑛
$
𝑧"⁄# is the margin of error.
%
 Confidence interval for the population proportion

• A 100 1 − 𝛼 % confidence interval for the population proportion

𝑝 is computed as
𝑝(1
̅ − 𝑝)̅
𝑝̅ ± 𝑧! ⁄"
𝑛

• Note that 𝑧! ⁄" is the value associated with the upper tail of the
standard normal distribution.
 Estimation
• While it is common to report the 95% confidence interval, in
theory we can construct an interval of any level of confidence.
• 𝛼 (alpha) denote the allowed probability of error
• The probability the estimation procedure will generate an interval that
does not contain the population proportion / mean in the long run
• Related to the significance level later in the hypothesis
• The confidence coefficient (degree of confidence,
confidence level%) (1 − 𝛼 ) is interpreted as the probability that
the estimation procedure will generate an interval that contains
the true population parameter.
 Critical values of Z

When sampling from the same population, using a fixed sample size, the higher the
confidence level 1 − 𝛼 , the wider confidence interval. Using a fixed confidence level,
the larger the sample size, n, the narrower the confidence interval.
The width of a confidence interval can be reduced only at the price of: a lower
level of confidence, or a larger sample.
 Example
Population consists of the Fortune 500 Companies (Fortune Web Site), as ranked by
Revenues. You are trying to to find out the average Revenues for the companies on the list.
The population standard deviation is $15,056.37. A random sample of 30 companies obtains
a sample mean of $10,672.87. Give a 95% and 90% confidence interval for the average
Revenues.

A 95% and 90% confidence interval for the population mean 𝜇 when 𝜎 is known is
! !
𝑥̅ ± 1.96
"
and 𝑥̅ ± 1.645 "
𝑛 = 30 95% CI 90% CI
𝜎 = 15,056.37 15056.37 15056.37
𝑥̅ = 10672.87 10672.87 ± 1.96 10672.87 ± 1.645
30 30
Example
Example: (Take home)
In a survey of 25 firms, seven of them experienced a cyber-attack
in the past six months. Construct a 90% confidence interval for the
population proportion of all firms that have experienced a cyber-
attack in the past six months.

𝑝̅ = 7⁄25 = 0.28
The normality condition is satisfied since 𝑛𝑝 ≥ 5 and 𝑛(1 − 𝑝) ≥ 5
For a 90% confidence interval, 𝛼 = 0.10 and 𝛼 ⁄2 = 0.05, 𝑧"⁄# =
𝑧$.$& =1.645
̅
((*+ ()̅ $.#.(*+$.#.)
𝑝̅ ± 𝑧"⁄# -
= 0.28 ± 1.645 #&
With 90% confidence, the percentage of firms that have experienced a
cyber-attack in the past size months is between between 13.2% and 42.8%.
Confidence interval for mean when 𝜎 is unknown
sample size is large, n > 30.
A 100 1 − 𝛼 % confidence interval for the population
mean 𝜇 when 𝜎 is unknown is
𝒔
𝑥̅ ± 𝑧;
< 𝑛
Confidence interval for mean when 𝜎 is unknown
and sample size is small < 30.
A 100 1 − 𝛼 % confidence interval for the population
mean 𝜇 when 𝜎 is unknown is
𝑠
𝑥̅ ± 𝑡;,=>
< 𝑛
• Note 𝑠 is the sample standard deviation and 𝑑𝑓 = 𝑛 − 1.
• 𝑡/,%& is the value of the t distribution with 𝑑𝑓 = 𝑛 − 1 that
0
;
cuts off the tails areas.
<
Example
A stock market analyst wants to estimate the average return on a certain
stock. A random sample of 15 days yields an average (annualized) return
of x = 10.37% and a standard deviation of s = 3.5%. Assuming a normal
population of returns, give a 95% confidence interval for the average
return on this stock.
A 95% confidence interval for the population mean 𝜇 when 𝜎 is
unknown and n<30 is
𝑠
𝑛 = 15 𝑥̅ ± 𝑡#,&'
$ 𝑛
𝑠 = 3.5
𝑥̅ = 10.37 95% CI
𝑡!,$% = 2.145 3.5
" 10.37 ± 2.145
15
CI of Two Means
independent and large sample independent and small sample

𝑦/. − 𝑦/" ± 𝑧! 𝑆𝐷(𝑦/. − 𝑦/" ) ( !" ! ! # ) ± "$%! " #$ & !" ! ! # '(
"

!"# #! " # " $ = $%& # #! $ + $%& # # " $ !"! !!!

"# # $" ! $ ! $ = + %
" " %" %!
# !! $ # ! " $
= % +
% D && %% D &&
' !( ' " (
! !" ! ""
= +
D! D"
Same population and small sample
The Pooled t-Test
$ ' " ! ' # % ± E !"
!
" )* #$$%&! $ ' " ! ' # %&

! !
!"##$%& !"##$%& ' '
"#"##$%& ( $' ! $ ! ) = + = !"##$%& +
%' %! %' %! #$ %&!! ! !'(&!" ! !'

(!" ! ")""! + (!! ! ")"!!

!
"#$$%&' =
(!" ! ") + (!! ! ")
Confidence interval for the population proportion
• The parameter p represents the proportion of successes in the
population.
• Use the sample proportion 𝑃/ as the point estimator of the population
proportion p.
• 𝑃/ is approximately normally distributed when 𝑛𝑝 ≥ 5 and
𝑛(1 − 𝑝) ≥ 5.
8(:;8)
• 𝐸 𝑃1 = 𝑝 and 𝑠𝑒 𝑃1 = "
8(:;8) ̅
8(:; 8)̅
• Replace "
with "
since 𝑝 is unknown.
Confidence Interval for the Difference
Between Two Proportions
How to develop a confidence interval for the difference between the
proportions of two populations?
1) Take a sample of size n1 from the first population
2) Take a sample of size n2 from the second population
3) Calculate the standard error of proportion for the first sample
4) Calculation the standard error of proportion for the first sample
Confidence Interval for the Difference
Between Two Proportions
Use the formula below to calculate the confidence interval
for the difference between two populations:

Confidence Interval for the Difference Between Two Proportions

The confidence interval for the difference between two proportions is

% !$" ! !$ # & ± " ! " #$ % !$" ! !$ # &'

where z* is the critical value and
!#!"#! !# ""#"
#$ $ !#! ! !# " % = + &
%! %"
 Interpretation of Confidence interval
• A 100 1 − 𝛼 % confidence interval for the population
Confidence intervals are often misinterpreted. For
example, a 90% confidence interval does not imply that
the probability the mean falls in the interval is 0.90.

• A 90% confidence interval implies 90% of the

intervals constructed will contain the mean.
 The Chi-Square (𝝌𝟐𝒓 ) Distribution
•The chi-square distribution is the probability distribution of
the sum of 𝑛 independent, squared standard normal random
? " A! ;B ? " ; : C"
variables 𝜒=>";: = ∑@>: = .
! !"

•The chi-square random variable cannot

be negative, is skewed to the right.

• The chi-square distribution

approaches a normal as the degrees of
freedom increase.

•The mean of the chi-square distribution

is equal to the degrees of freedom
parameter, 𝐸 𝜒&" = 𝑟. The variance of a
chi-square is equal to twice the number
of degrees of freedom, Var 𝜒&" = 2𝑟.
 Confidence Interval for the
Population Variance
A 100 1 − 𝛼 % confidence interval for the population
variance 𝜒 ! (Population is assumed normal) is :
𝑛 − 1 𝑠" 𝑛 − 1 𝑠"
" ,
𝜒!/" 𝜒." 1!/"
!
where 𝜒"/! is the value of the chi-square distribution with n - 1
" !
degrees of freedom that cuts off an area to its right and 𝜒$%"/!
!
"
is the value of the distribution that cuts off an area of to its left
!
"
(equivalently, an area of 1 − to its right)
!

Because the chi-square distribution is skewed, the confidence

interval for the population variance is not symmetric.
Example
In an automated process, a machine fills cans of coffee. If the average amount
filled is different from what it should be, the machine may be adjusted to
correct the mean. If the variance of the filling process is too high, however, the
machine is out of control and needs to be repaired. Therefore, from time to
time regular checks of the variance of the filling process are made. This is
done by randomly sampling filled cans, measuring their amounts, and
computing the sample variance. A random sample of 30 cans gives an estimate
𝑠 $ = 18,540. Give a 95% confidence interval for the population variance, 𝜎 $ .
Sample-Size Determination
Before determining the necessary sample size, the researcher must answer three questions:

•How close do you want your sample estimate to be to the unknown parameter is called the
!
the desired bound or margin of error, or the sampling error 𝐵 = 𝑧#/$ .
√"

•What do you want the desired confidence level (1 − 𝛼) to be so that the distance between
your estimate and the parameter is less than or equal to B?

•What is your estimate of the variance (or standard deviation) of the population in question?
Sample-Size Determination
Minimum required sample size in estimating the population mean,

"
𝑧!/" 𝜎
𝑛≥
𝐵"

Minimum required sample size in estimating the population proportion,

"
𝑧!/" 𝑝 (1 − 𝑝)
𝑛≥
𝐵"
Example
A marketing research firm wants to conduct a survey to estimate
the average amount spent on entertainment by each person
visiting a popular resort. The people who plan the survey would
like to determine the average amount spent by all people visiting
the resort to within $120, with 95% confidence. From past
operation of the resort, an estimate of the population standard
deviation is s = $400. What is the minimum required sample size?

"
𝑧! 𝜎 1.96 × 400 "
"
𝑛≥ = = 42.68
𝐵" 120"
 Example
The manufacturers of a sports car want to estimate the proportion
of people in a given income bracket who are interested in the
model. The company wants to know the population proportion, p,
to within 0.01 with 99% confidence. Current company records
indicate that the proportion p may be around 0.25. What is the
minimum required sample size for this survey?

𝑧!" 𝑝 1 − 𝑝
" 2.576" 0.25 0.75
𝑛≥ = = 124.42
𝐵" 0.1"