ENGINEERING DATA ANALYSIS

Hand-out No. 4: Statistical Interval

Hand out No. 4 STATISTICAL INTERVAL Time Frame: 2 hours

-

Learning Objectives
• Construct confidence intervals on the mean of a normal distribution, using either the
normal distribution or the t-distribution method
• Explain the three types of interval estimates: confidence intervals, prediction intervals,
and tolerance intervals
• Use the general method for constructing a confidence interval
Content Outline
1. Confidence Intervals
a. Single Sample
b. Multiple Samples
2. Prediction Intervals
3. Tolerance Intervals

STATISTICAL INTERVAL
A Statistical Interval provide a way to quantify the uncertainty about a scalar quantity
of a population based on a random sample of data. A statistical interval is a range, comprising
a lower bound and an upper bound, which is expected to contain the true population value, with
a certain degree of confidence.

CONFIDENCE INTERVAL
An interval estimate for a population parameter is called a confidence interval. A
confidence interval is the mean of your estimate plus and minus the variation in that estimate.
This is the range of values you expect your estimate to fall between if you redo your test, within
a certain level of confidence. Confidence, in statistics, is another way to describe probability.

CONFIDENCE INTERVAL: SINGLE SAMPLE

Confidence Interval On the Mean of a Normal Distribution, Variance Known
The basic ideas of a confidence interval (CI) are most easily understood by initially considering
a simple situation. Suppose that we have a normal population with unknown mean and known
variance σ2. This is a somewhat unrealistic scenario because typically both the mean and
variance are unknown. However, in subsequent sections we will present confidence intervals
for more general situations.

Page 1 of 13
Mechanical Engineering Department, College of Engineering and Architecture
ENGINEERING DATA ANALYSIS
Hand-out No. 4: Statistical Interval

If x̄ is the sample mean of a sample mean of a random sample of size n from a normal
population with known variance σ2 , a 100(1 ─ α)% CI on µ is given by

𝜎 𝜎
X̅ − zα⁄2 ≤µ≤̅
X + zα⁄2
√n √n

Equation 4.1

where zα/2 is the z-value leaving an area of α/2 to the right.

EXAMPLE 4.1
The average zinc concentration recovered from a sample of measurements taken in 36
different locations in a river is found to be 2.6 grams per milliliter. Find the 95% and 99%
confidence intervals for the mean zinc concentration in the river. Assume that the population
standard deviation is 0.3 gram per milliliter.

Given:
x̄ = 2.6
𝛔 = 0.3
n = 36
α = 95%, 99%
In CI= 95%

Solving for 𝑍 ,

1+𝛼
𝑍 =
2
Equation 4.2

1 + 0.95
𝑍 = = 0.975 = 1.96
2
𝜎 𝜎
̅ − zα⁄2
X ̅ + zα⁄2
≤µ≤X
√n √n
0.3 0.3
2.6 − 1.96 ( ) ≤ µ ≤ 2.6 + 1.96 ( )
√36 √36
𝟐. 𝟓𝟎𝟐 ≤ µ ≤ 𝟐. 𝟔𝟗𝟖

Page 2 of 13
Mechanical Engineering Department, College of Engineering and Architecture
ENGINEERING DATA ANALYSIS
Hand-out No. 4: Statistical Interval

In CI= 99%

1 + 0.99
𝑍 = = 0.995 = 2.58
2

𝜎 𝜎
̅
X − zα⁄2 ≤µ≤̅
X + zα⁄2
√n √n
0.3 0.3
2.6 − 2.58 ( ) ≤ µ ≤ 2.6 + 2.58 ( )
√36 √36

𝟐. 𝟒𝟕 ≤ µ ≤ 𝟐. 𝟕𝟑

Based on the given, the confidence intervals for the mean zinc concentration in the river at
95% is 2.502 < µ < 2.698 and at 99% is 2.47 < µ < 2.73.

Z- score table

Page 3 of 13
Mechanical Engineering Department, College of Engineering and Architecture
ENGINEERING DATA ANALYSIS
Hand-out No. 4: Statistical Interval

Sample Size for Specified Error on the Mean, Variance Known

If x̄ is used as an estimate of μ, we can be 100(1 − α)% confident that the error, e, will not
exceed

𝜎
𝐸 ≤ 𝑍𝛼/2
√𝑛
Equation 4.3

If x̄ is used as an estimate of µ, we can be 100(1 ─ α)% confident that the error | x̄ ─ µ |

will not exceed a specified amount e when the sample size is

zα⁄2 𝜎 2
𝑛=( )
𝐸
Equation 4.4

EXAMPLE 4.2
How large a sample is required if we want to be 95% confident that our estimate of μ in
Example 4.1 is off by less than 0.05?

Given:
x̄=2.6
𝛔 = 0.3
a = 95%
z = 1.96
E = 0.05

Solve For n:

zα⁄2 𝜎 2
𝑛=( )
𝐸

1.96(0.3)
𝑛=( )²
0.05

𝑛 = 11.76
𝑛 = 138.30 or
𝒏 = 𝟏𝟑𝟗

Therefore, we can be 95% confident that a random sample of size 139 will provide
an estimate x̄ differing from μ by an amount less than 0.05.

Page 4 of 13
Mechanical Engineering Department, College of Engineering and Architecture
ENGINEERING DATA ANALYSIS
Hand-out No. 4: Statistical Interval

Notice the general relationship between sample size, desired length of the confidence
interval 2E, confidence level 100(1─α), and standard deviation 𝜎:

• As the desired length of the interval 2E decreases, the required sample size 𝑛 increases
for a fixed value of 𝜎 and specified confidence.
• As 𝜎 increases, the required sample size 𝑛 increases for a fixed desired length 2E and
specified confidence.
• As the level of confidence increases, the required sample size 𝑛 increases for fixed
desired length 2E and standard deviation 𝜎.

One-Sided Confidence Bounds on the Mean, Variance Known

If x̄ is the mean of a random sample of size n from a population with variance σ2, the one-
sided 100(1 − α)% confidence bounds for μ are given by

Upper one-sided bound:

σ
x̄ + 𝑧α
√𝑛
Equation 4.5
Lower one-sided bound:
σ
x̄ − 𝑧α
√𝑛
Equation 4.6

EXAMPLE 4.3
In a psychological testing experiment, 25 subjects are selected randomly and their reaction
time, in seconds, to a particular stimulus is measured. Past experience suggests that the
variance in reaction times to these types of stimuli is 4 sec 2 and that the distribution of
reaction times is approximately normal. The average time for the subjects is 6.2 seconds.
Give an upper 95% bound for the mean reaction time.

Given:
x̄ = 2.6
𝛔2 = 4
α = 95%
n = 25

Page 5 of 13
Mechanical Engineering Department, College of Engineering and Architecture
ENGINEERING DATA ANALYSIS
Hand-out No. 4: Statistical Interval

Solve For Upper One-sided Bound:

σ
x̄ + 𝑧α
√𝑛
√4
2.6 + 1.645
√25
=𝟔. 𝟖𝟓𝟖

T-Distribution

Let X1 , X2 , … , Xn be a random sample from a normal distribution with unknown mean

μ and unknown variance σ 2 . The random variable

x̄ − μ
𝑇=
𝑠/√𝑛
Equation 4.7
e
has a t distribution with n - 1 degrees of freedom

ONE-SIDED CONFIDENCE BOUNDS

The confidence intervals and resulting confidence bounds discussed thus far are two-sided (i.e.,
both upper and lower bounds are given). However, there are many applications in which only
one bound is sought. For example, if the measurement of interest is tensile strength, the
engineer receives better information from a lower bound only. This bound communicates the
worst-case scenario. On the other hand, if the measurement is something for which a relatively
large value of μ is not profitable or desirable, then an upper confidence bound is of interest. An
example would be a case in which inferences need to be made concerning the mean mercury
composition in a river. An upper bound is very informative in this case. One-sided confidence
bounds are developed in the same fashion as two-sided intervals.
One-Sided Confidence Bounds on the Mean, Variance Known

If x̄ and s are the mean and standard deviation of a random sample from a normal
population with unknown variance σ 2, a 100(1−α)% confidence interval for μ is
σ σ
x̄ − 𝑡α⁄2 < μ < x̄ + 𝑡α⁄2
√𝑛 √𝑛
Equation 4.8
where tα/2 is the t-value with v = n − 1 degrees of freedom, leaving an area of α/2 to the right.

Page 6 of 13
Mechanical Engineering Department, College of Engineering and Architecture
ENGINEERING DATA ANALYSIS
Hand-out No. 4: Statistical Interval

EXAMPLE 4.4
The contents of seven similar containers of sulfuric acid are 9.8, 10.2, 10.4, 9.8, 10.0, 10.2,
and 9.6 liters. Find a 95% confidence interval for the mean contents of all such containers,
assuming an approximately normal distribution.

Given:
α=95%
Samples = 9.8, 10.2, 10.4, 9.8, 10, 10.2, 9.6
n=7

First solve for x̄ :

9.8 + 10.2 + 10.4 + 9.8 + 10 + 10.2 + 9.6

x̄ =
7
x̄ = 10

Solve for S:

(9.8 − 10)² + (10.2 − 10)² + (10.4 − 10)² + (9.8 − 10)² + (10 − 10)² + (10.2 − 10)² + (9.6 − 10)²
𝑆² =
7

0.48
𝑆² =
7
𝑆² = 0.068571428571428
𝑆 = √0.068571428571428
𝑆 = 0.26186146828319

𝑣=𝑛−1
𝑣 =7−1
𝑣=6

Solve for t:
1−α
𝑡=
2
1 − 0.95
𝑡=
2
𝑡 = 0.025
𝑡 = 2.4469

Solve for Confidence Interval:

σ σ
x̄ − 𝑡α⁄2 < μ < x̄ + 𝑡α⁄2
√𝑛 √𝑛

0.2618 0.2618
10 − 2.4469 < μ < 10 + 2.4469
√7 √7

10 − .2422 < μ < 10 + .2422

𝟗. 𝟕𝟓 < 𝛍 < 𝟏𝟎. 𝟐𝟒

Page 7 of 13
Mechanical Engineering Department, College of Engineering and Architecture
ENGINEERING DATA ANALYSIS
Hand-out No. 4: Statistical Interval

STUDENT’S T-DISTRIBUTION TABLE

CONFIDENCE INTERVAL: MULTIPLE SAMPLE

Often statisticians recommend that even when normality cannot be assumed, 𝝈 is unknown,
and n ≥ 30, 𝒔 can replace 𝝈 and the confidence interval

s
X̅ ± zα⁄2
√n
Equation 4.9
Where:

X̅ = Sample mean
zα⁄2 = Z score value
s = Standard deviation
n = Sample size

may be used. This is often referred to as a large-sample confidence interval. The

justification lies only in the presumption that with a sample as large as 30 and the population
distribution not too skewed, s will be very close to the true 𝝈 and thus the Central Limit
Theorem prevails. It should be emphasized that this is only an approximation and the quality
of the result becomes better as the sample size grows larger.
Page 8 of 13
Mechanical Engineering Department, College of Engineering and Architecture
ENGINEERING DATA ANALYSIS
Hand-out No. 4: Statistical Interval

EXAMPLE 4.5
Scholastic Aptitude Test (SAT) mathematics scores of a random sample of 500 high school
seniors in the state of Texas are collected, and the sample mean and standard deviation are
found to be 501 and 112, respectively. Find a 99% confidence interval on the mean SAT
mathematics score for seniors in the state of Texas.

Given:
𝑛 = 500
̅ = 501
𝑋̅
𝑠 = 112
𝛼 = 99%

Solving for Z score;

1+𝛼
𝑍 =
2

1 + .99
𝑍 = = 0.995 = 2.58
2

Solving for confidence interval;

𝑠
𝑋̅ ± 𝑧𝛼⁄2
√𝑛
𝑠 𝑠
𝑋̅ − 𝑧𝛼⁄2 ≤ 𝜇 ≤ 𝑋̅ + 𝑧𝛼⁄2
√𝑛 √𝑛

112 112
501 − 2.58 ≤ 𝜇 ≤ 501 + 2.58
√500 √500

𝟒𝟖𝟖. 𝟏 ≤ 𝝁 ≤ 𝟓𝟏𝟑. 𝟗

PREDICTION INTERVAL
Sometimes, other than the population mean, the experimenter may also be interested in
predicting the possible value of a future observation. For instance, in quality control,
the experimenter may need to use the observed data to predict a new observation. A process
that produces a metal part may be evaluated on the basis of whether the part meets
specifications on tensile strength. On certain occasions, a customer may be interested in
purchasing a single part. In this case, a confidence interval on the mean tensile strength does
not capture the required information. The customer requires a statement regarding the
uncertainty of a single observation. This type of requirement is nicely fulfilled by the
construction of a prediction interval. It is quite simple to obtain a prediction interval for the
situations we have considered so far. Assume that the random sample comes from a normal
population with unknown mean μ and known variance σ2. It is known, that the variance of 𝑋̅ is
σ2/n. However, to predict a new observation, not only do we need to account for the variation
due to estimating the mean, but also we should account for the variation of a future
observation.

Page 9 of 13
Mechanical Engineering Department, College of Engineering and Architecture
ENGINEERING DATA ANALYSIS
Hand-out No. 4: Statistical Interval

Prediction Interval of a Future Observation, σ2 Known

For a normal distribution of measurements with unknown mean µ and unknown variance
σ2, a 100(1 − α)% prediction interval of a future observation x0 is

X̅ − 𝑧α⁄2 𝑠√1 + 1/𝑛 ≤ 𝑥0 ≤ 𝑋̅ + 𝑧𝛼⁄2 𝑠√1 + 1/𝑛

Equation 4.10

where zα/2 is the z-value leaving an area of α/2 to the right.

EXAMPLE 4.6
Due to the decrease in interest rates, the First Citizens Bank received a lot of mortgage
applications. A recent sample of 50 mortgage loans resulted in an average loan amount of
$257,300. Assume a population standard deviation of $25,000. For the next customer who
fills out a mortgage application, find a 95% prediction interval for the loan amount.

Given:
n= 50
X̅ = $257,300
S= $25,000
α= 95%

Solve for Z
1 + 0.95
𝑍 = = 0.975 = 1.96
2

Solve for Prediction Interval

X̅ − 𝑧α⁄2 𝑠√1 + 1/𝑛 ≤ 𝑥0 ≤ 𝑋̅ + 𝑧𝛼⁄2 𝑠√1 + 1/𝑛

$257,300 − 1.96( $25,000)√1 + 1/50 ≤ 𝑥0 ≤ $257,300 + 1.96($25,000)√1 + 1/50

$𝟐𝟎𝟕, 𝟖𝟏𝟐. 𝟒𝟑 ≤ 𝒙𝟎 ≤ $𝟑𝟎𝟔, 𝟕𝟖𝟕. 𝟓𝟕

Page 10 of 13
Mechanical Engineering Department, College of Engineering and Architecture
ENGINEERING DATA ANALYSIS
Hand-out No. 4: Statistical Interval

For a normal distribution of measurements with unknown mean μ and unknown variance
σ2, a 100(1 − α)% prediction interval of a future observation x 0 is

X̅ − 𝑡α⁄2 𝑠√1 + 1/𝑛 ≤ 𝑥0 ≤ X̅ + 𝑡𝛼⁄2 𝑠√1 + 1/𝑛

Equation 4.11

where tα/2 is the t-value with v = n − 1 degrees of freedom, leaving an area of

α/2 to the right.

One-sided prediction intervals can also be constructed. Upper prediction bounds apply in cases
where focus must be placed on future large observations. Concern over future small
observations calls for the use of lower prediction bounds. The upper bound is given by

𝑋̅ + 𝑡𝛼⁄2 𝑠√1 + 1/𝑛

and the lower bound by

X̅ − 𝑡α⁄2 𝑠√1 + 1/𝑛

EXAMPLE 4.7
A meat inspector has randomly selected 30 packs of 95% lean beef. The sample resulted in
a mean of 96.2% with a sample standard deviation of 0.8%. Find a 99% prediction interval
for the leanness of a new pack.

Given:
n= 30
X̅= 96.2%
S= 0.8%
α= 99%

𝑣 = 𝑛 − 1 = 30 − 1 = 29
Solve for t
1 − 0.99
𝑡 = = 0.005 = 2.7564
2

Solve for Prediction Interval

X̅ − 𝑡α⁄2 𝑠√1 + 1/𝑛 ≤ 𝑥0 ≤ 𝑋̅ + 𝑡𝛼⁄2 𝑠√1 + 1/𝑛

96.2 − 2.7564( 0.8)√1 + 1/30 ≤ 𝑥0 ≤ 96.2 + 2.7564(0.8)√1 + 1/30

𝟗𝟑. 𝟗𝟔 ≤ 𝒙𝟎 ≤ 𝟗𝟖. 𝟒𝟒

Page 11 of 13
Mechanical Engineering Department, College of Engineering and Architecture
ENGINEERING DATA ANALYSIS
Hand-out No. 4: Statistical Interval

TOLERANCE INTERVAL
If process specifications are important (e.g., manufacturing), the manager of the process is
concerned about long-range performance, not the next observation. One must attempt to
determine bounds that, in some probabilistic sense, “cover” values in the population (i.e.,
the measured values of the dimension).
One method of establishing the desired bounds is to determine a confidence interval on a fixed
proportion of the measurements. This is best motivated by visualizing a situation in which we
are doing random sampling from a normal distribution with known mean μ and variance σ2.
Clearly, a bound that covers the middle 95% of the population of observations is

μ ± 1.96σ.
This is called a tolerance interval, and indeed its coverage of 95% of measured observations
is exact.

However, in practice, μ and σ are seldom known; thus, the user must apply

X̅ ± ks.
Equation 4.12

where k is determined such that one can assert with 100(1 − γ)% confidence that the
given limits contain at least the proportion 1 − α of the measurements.

EXAMPLE 4.8
A foundry has sampled hardness of 15-cylinder head casting. Cylinder head is a typical
component used in diesel engines. The quality engineer finds that the average hardness is
230 Brinell Hardness Number (BHN) and standard deviation is 20 BHN. Assuming normal
distribution, estimate two-sided tolerance limits such that 99% of the distribution will lie in
these limits with probability of 0.95.

Given:
n= 15
X̅ (Average)= 230
S (Standard Deviation)= 20
CL (Confidence Level)= 95% or 0.95
Proportion= 99% or 0.99

Formula:
X̅ ± ks
Solution:

X̅ ± ks
̅
X − ks , ̅X + ks
230 − (3.878)(20) , 230 + (3.878)(20)

Page 12 of 13
Mechanical Engineering Department, College of Engineering and Architecture
ENGINEERING DATA ANALYSIS
Hand-out No. 4: Statistical Interval

152.44 , 307.56

Therefore, we can say that with 95% confidence that 99% of the cylinder heads will have
hardness between 152 and 308.

FACTORS FOR TOLERANCE INTERVALS TABLE

Keep the purpose of the three types of interval estimates clear:

• A confidence interval bounds population or distribution parameters (such as mean
viscosity)
• A tolerance interval bounds a selected proportion of a distribution.
• A prediction interval bounds future observations from the population or distribution.

Page 13 of 13
Mechanical Engineering Department, College of Engineering and Architecture