Nguyễn Ngọc Tứ

Lecture 5

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 1 / 18

Lecture outline

ˆ Point Estimate
ˆ Confidence Intervals
ˆ Confidence Intervals for a Population Mean and Proportion
ˆ One-Sided Confidence Intervals
ˆ The One-Sample t Confidence Interval

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 2 / 18

Point estimate
Definition. A point estimate of a parameter θ is a single number that can be regarded as a
sensible value for θ. A point estimate is obtained by selecting a suitable statistic and computing
its value from the given sample data. The selected statistic is called the point estimator of θ.
Definition. A point estimator θ̂ is said to be an unbiased estimator of θ if E (θ̂) = θ, for every
possible value of θ. If θ̂ is not unbiased, the difference E (θ̂) − θ is called the bias of θ̂.
Proposition.
1. When X ∼ Bin(n, p), the sample proportion p̂ = X /n is an unbiased estimator of p.
2. Let X1 , . . . , Xn be a random sample from a distribution with mean µ and variance σ 2 .
Then the estimator X is unbiased for estimating µ and

(Xi − X )2
P
2 2
σ̂ = S =
n−1

is unbiased for estimating σ 2 .

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 3 / 18
Confidence Interval
Definition. A 100(1 − α)% confidence interval for the mean µ of a normal population when
the value of σ is known is given by
 
σ σ σ
x − zα/2 · √ , x + zα/2 · √ = x ± zα/2 · √
n n n
A confident level γ := 100(1 − α)% ⇒ Φ(zα/2 ) = 1 − α/2 ⇒ zα/2 = · · ·
γ 90% 91% 92% 93% 94% 95% 96% 97% 98% 99%
zα/2 1.645 1.695 1.751 1.812 1.881 1.96 2.054 2.170 2.326 2.576

Example 1. Let X ∼ N(µ, σ 2 ). The quantities needed for computation of the 95% CI for true
average preferred height are σ = 2, n = 31, and x = 80.

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 4 / 18

Confidence Interval
Definition. A 100(1 − α)% confidence interval for the mean µ of a normal population when
the value of σ is known is given by
 
σ σ σ
x − zα/2 · √ , x + zα/2 · √ = x ± zα/2 · √
n n n
A confident level γ := 100(1 − α)% ⇒ Φ(zα/2 ) = 1 − α/2 ⇒ zα/2 = · · ·
γ 90% 91% 92% 93% 94% 95% 96% 97% 98% 99%
zα/2 1.645 1.695 1.751 1.812 1.881 1.96 2.054 2.170 2.326 2.576

Example 1. Let X ∼ N(µ, σ 2 ). The quantities needed for computation of the 95% CI for true
average preferred height are σ = 2, n = 31, and x = 80.
With 95% CI, we have α = 0.05 and zα/2 = z0.025 = 1.96. Therefore, the resulting interval is
σ 2
x ± zα/2 · √ = 80 ± 1.96 √ = (79.2959, 80.7041)
n 31
Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 4 / 18
Confidence Interval - Sample Size
Example 2. Assume that the helium porosity (in percentage) of coal samples taken from any
particular seam is normally distributed with true standard deviation 0.75.
a. Compute a 95% CI for the true average porosity of a certain seam if the average porosity for
20 specimens from the seam was 4.85.
b. Compute a 98% CI for true average porosity of another seam based on 16 specimens with a
sample average porosity of 4.56.
c. How large a sample size is necessary if the width of the 95% interval is to be 0.40?
 
σ σ
Solution. The sample size necessary for the CI x − zα/2 · √ , x + zα/2 · √ to have a width
n n
σ  σ 2

w = 2zα/2 · √ is n = 2zα/2 · .
n w
0.75 2
 
Therefore, we have n = 2 · 1.96 = 54.0225. Since n must be integer, a sample size of
0.4
55 is required.
Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 5 / 18
Large-Sample Confidence Intervals for a Population Mean (n > 40)

Let X1 , . . . , Xn be a random sample from a population having a mean µ and standard deviation
σ. Provided that n is large, the Central Limit Theorem (CLT) implies that X has approximately
a normal distribution whatever the nature of the population distribution. Then

X −µ
Z= √ → N(0, 1)
S/ n

This implies that  

s s
x − zα/2 · √ , x + zα/2 · √
n n
is a large-sample confidence interval for µ with confidence level approximately 100(1 − α)%.
This formula is valid regardless of the shape of the population distribution.

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 6 / 18

Large-Sample Confidence Intervals for a Population Mean (n > 40)

Example 1. Measuring the diameter X (unit: mm) of a type of machine manufactured by

company M obtains the following table:

X 87 89 91 93 95 97 99
The number of types of machine 37 45 69 83 71 45 32

Calculate a 96% CI for true average diameter of types of machine, knowing that X has a normal
distribution.
Solution. Sample mean: x = . . . . . ., n = . . . . . ., s = . . . . . .
Let µ be true average diameter of types of machine.
With 96% CI: γ = 96% ⇒ zα/2 = . . . . . . . . .
The interval is
s
x ± zα/2 · √ = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
n

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 7 / 18

Calculator fx for statistics
Calculator fx 570 for statistics
1. Shift → Mode → ⇓ 4 → 1. On
2. Mode → 3 : STAT → 1.
X FREQ
87 37
Input data, 89 45 then press AC.
.. ..
. .
99 32
3. Shift → 1 → 4: Var → 1. n Shift → 1 → 4: Var → 2. x
Shift → 1 → 4: Var → 4. xσn−1 or s.
Calculator fx 580 VNX for statistics
1. Shift → Mode → ⇓ 3 → 1. On
2. Mode → 6 → 1. Input data, then press AC.
3. OPTN → 2 → n, x, sx = s
Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 8 / 18
Large-Sample Confidence Intervals for a Population Mean (n > 40)

Example 2. The weight X (kg) of pigs selected randomly on a farm is given in the following
table
X 65-85 85-95 95-105 105-115 115-135
The number of pigs 8 48 66 42 10

Calculate a 90% CI for true average weight of the pig, knowing that X has a normal distribution.
Example 3. Expenditure X (million/month) of some people randomly selected in region A:
X 3.2-3.7 3.7-4.2 4.2-4.7 4.7-5.2 5.2-5.7 5.7-6.2 6.2-6.7
ni 23 33 55 79 45 29 18

Suppose that X has a normal distribution. Calculate a 98% CI for the population mean of the
expenditure.

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 9 / 18

Small-Sample Confidence Intervals for a Population Mean (n ≤ 40)

When X is the mean of a random sample of size n (n ≤ 40) from a normal distribution with
mean µ, the random variable
X −µ
T = √
s/ n
has a t distribution with n − 1 degrees of freedom (df).
Let x and s be the sample mean and sample standard deviation computed from the results of
a random sample from a normal population with mean µ. Then a 100(1 − α)% confidence
interval for µ is
 
s s s
x − tα/2,n−1 · √ , x + tα/2,n−1 · √ = x ± tα/2,n−1 · √ .
n n n

where tα/2,n−1 is called a t critical value.

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 10 / 18

Small-Sample Confidence Intervals for a Population Mean (n ≤ 40)
Example 1. Consider the following sample of fat content (in percentage) of n = 10 randomly
selected hot dogs:

25.2 21.3 22.8 17.0 29.8 21.0 25.5 16.0 20.9 19.5

A 95% CI for the population mean fat content is

s 4.1341
x ± t0.025,9 . √ = 21.90 ± 2.262 × √ = (18.8141, 24.9859)
n 10
Ex 37, p.293. A study of the ability of individuals to walk in a straight line reported the
accompanying date on cadence for a sample of n = 20 randomly selected healthy men.
.95 .85 .92 .95 .93 .86 1.00 .92 .85 .81
.78 .93 .93 1.05 .93 1.06 1.06 .96 .81 .96

Calulate and interpret a 95% confidence interval for population mean cadence.
Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 11 / 18
Small-Sample Confidence Intervals for a Population Mean (n ≤ 40)

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 12 / 18

A Confidence Interval for a Population Proportion
Let p denote the proportion of “successes” in a population, where success identifies an individual
or object that has a specified property. A random sample of n individuals is to be selected,
and Xpis the number of successes in the sample. Then X ∼ Bin(n, p) with EX = np and
σX = np(1 − p).
If both np ≥ 10 and n(1 − p) ≥ 10, X has approximately a normal distribution.
X
The natural estimator of p is p̂ = . Then
n
!
p̂ − p
P −zα/2 < p < zα/2 ≃ 1 − α
p(1 − p)/n
and a confidence interval for a population proportion p with confidence level approximately
100(1 − α)% is r r !
p̂(1 − p̂) p̂(1 − p̂)
p ∈ p̂ − zα/2 , p̂ + zα/2
n n
Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 13 / 18
A Confidence Interval for a Population Proportion
Example 1.The article “Limited Yield Estimation for Visual Defect Sources” reported that, in
a study of a particular wafer inspection process, 356 dies were examined by an inspection probe
and 201 of these passed the probe. Assuming a stable process, calculate a 95% confidence
interval for the proportion of all dies that pass the probe.
Solution.
Let p be the true proportion of all dies that pass the probe.
201
Sample proportion p̂ = = ......
356
With 95% CI: γ = 95% ⇒ zα/2 = . . . . . . . . .
The interval is
r
p̂(1 − p̂)
p ∈ p̂ ± zα/2 = ..............................
n

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 14 / 18

A Confidence Interval for a Population Proportion
Example 2. An April 2009 survey of 2253 American adults conducted by the Pew Research
Center’s Internet & American Life Project revealed that 1262 of the respondents had at some
point used wireless means for online access.
a. Calculate and interpret a 95% CI for the proportion of all American adults who at the time
of the survey had used wireless means for online access.
b. What sample size is required if the desired width of the 95% CI is to be at most 0.04,
irrespective of the sample results?

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 15 / 18

One-Sided Confidence Intervals (Confidence Bounds)

The confidence intervals discussed thus far give both a lower confidence bound and an upper
confidence bound for the parameter being estimated. In some circumstances, an investigator
will want only one of these two types of bounds.

Two Sided CI Upper Lower

ESTIMATION
(One-Sided CI) (One-Sided CI)
σ σ σ
σ known x ± zα/2 · √ x + zα · √ x − zα · √
n n n
s s s
σ unknown, n > 40 x ± zα/2 · √ x + zα · √ x − zα · √
n n n
s s s
σ unknown, n ≤ 40 x ± tα/2,n−1 √ x + tα,n−1 √ x − tα,n−1 √
q n q n q n
p̂(1−p̂) p̂(1−p̂) p̂(1−p̂)
proportion p̂ ± zα/2 n p̂ + zα n p̂ − zα n

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 16 / 18

One-sided confidence bound
Some values of zα/2 and zα with γ := 100(1 − α)% CI.

γ 90% 91% 92% 93% 94% 95% 96% 97% 98% 99%
zα/2 1.645 1.695 1.751 1.812 1.881 1.96 2.054 2.170 2.326 2.576
zα 1.282 1.341 1.405 1.476 1.555 1.645 1.751 1.881 2.054 2.326

Example 1. Measuring the diameter X (unit: mm) of a type of machine manufactured by

company M obtains the following table:

X 87 89 91 93 95 97 99
The number of types of machine 37 45 69 83 71 45 32

a. Calculate an 98% upper confidence bound and an 98% lower confidence bound for the true
average diameter of types of machine.
b. Find an upper confidence bound and an lower confidence bound for the proportion of types
of a machine which diameter exceeds 97 mm at 94% the confidence level.
Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 17 / 18
One-sided confidence bound

Solution. a. Sample mean: x = 92.9319, n = 382, s = 3.4299.

With 98% CI: γ = 98% ⇒ zα =
s
The upper confidence bound for true average diameter of types of machine x+zα · √ = . . . . . . . . . . . . . . . . . . . .
n
s
The lower confidence bound for true average diameter of types of machine x−zα · √ = ....................
n
32
b. The proportion of types of a machine which diameter exceeds 97 mm is p̂ = .
382
With 94% CI: γ = 94% ⇒ zα =
The upper
r confidence bound for the proportion of types of a machine which diameter exceeds 97 mm
p̂(1 − p̂)
p̂ + zα = ..............................
n
The lower
r confidence bound for the proportion of types of a machine which diameter exceeds 97 mm
p̂(1 − p̂)
p̂ − zα = ..............................
n

Nguyễn Ngọc Tứ Lecture 5 2023 - 2024 18 / 18