Chapter 5: Probability and Statistics for Engineering and the

Sciences (J.L. Devore)

STA1006S 2024

1 / 21
Lecture Objectives

Chapter 5.4: pp.230 - 236

▶ Know and apply the rules about the distribution of the sample mean and sample
total for normal and non-normal cases.
▶ Define and apply the Central Limit Theorem.
▶ Know and apply the Central Limit Theorem for Binomial and Poisson
environments if the required criteria is met.

2 / 21
Homework discussion

Exercise 42, p.230

3 / 21
Homework discussion

Exercise 43, p.230

4 / 21
5.4: The distribution of the sample mean

▶ We rely on the properties of the sampling distribution of X̄ when drawing

conclusions about the population mean µ.
▶ Proposition (without proof):
Let X1 , X2 , ..., Xn be a random sample from a distribution with mean µ and
standard deviation σ. Then
1. E (X̄ ) = µX̄ = µ
2
2. V (X̄ ) = σX̄2 = σn and σX̄ = σ
√
n
.
▶ If T0 = X1 + X2 + ... + Xn (the sample total), then E (T0 ) = nµ, V (T0 ) = nσ 2 ,
√
and σT0 = nσ.

5 / 21
5.4: The distribution of the sample mean

What can we conclude from the proposition on the previous slide?

▶ Sampling distribution of X̄ is centered at the mean of the population from which
the sample has been selected.
▶ The sampling distribution of X̄ becomes more concentrated about µ as n
increases.
▶ The sampling distribution of T0 becomes more spread out as n increases.
▶ σX̄ = √σ = standard error of the mean (the magnitude of a typical deviation of
n
the sample mean from the population mean).

6 / 21
5.4: The distribution of the sample mean

Example 5.25
The expected number of cycles to first acoustic emission is µ = 28000 and the
standard deviation is σ = 5000.
Let X1 , X2 , ..., X25 be a random sample, where each Xi is the number of cycles on a
different randomly selected specimen.
1. Calculate E (X̄ ) and V (X̄ ).
2. Calculate E (T0 ) and σT0 .
3. How will the values calculated in (1.) and (2.) be affected if the sample size n
changes from 25 to 200?

7 / 21
5.4: The distribution of the sample mean
The case of a normal population distribution
Proposition (without proof):
Let X1 , X2 , ..., Xn be a random sample from a normal distribution with mean µ and
standard deviation σ. Then, for any sample size n:
2
▶ X̄ ∼ N(µ, σn )
▶ T0 ∼ N(nµ, nσ 2 ).
Probabilities P(a ≤ X̄ ≤ b) and P(a ≤ T0 ≤ b) can be calculated by simply
standardizing.

8 / 21
5.4: The distribution of the sample mean

Adaptation of Exercise 52, p. 237

The lifetime of a certain type of battery is normally distributed with a mean value of
15 hours and standard deviation of 2 hours. There are 8 batteries in a package.
▶ Calculate the probability that the average lifetime of the package is less than 17
hours.

9 / 21
5.4: The distribution of the sample mean

Adaptation of Exercise 52, p. 237 continued

The lifetime of a certain type of battery is normally distributed with a mean value of
15 hours and standard deviation of 2 hours. There are 8 batteries in a package.
▶ Calculate P(99 ≤ T0 < 104).

10 / 21
Adaptation of Exercise 52, p. 237 continued
The lifetime of a certain type of battery is normally distributed with a mean value of
15 hours and standard deviation of 2 hours. There are 8 batteries in a package.
▶ What lifetime value is such that the total lifetime of all batteries in a package
exceeds that value for only 5% of all packages?

11 / 21
5.4: The distribution of the sample mean
The central limit theorem (CLT)
Theorem (without proof):
Let X1 , X2 , ..., Xn be a random sample from a distribution with mean µ and variance
σ 2 . Then, if n is sufficiently large (Rule of thumb: n > 30):
▶ X̄ ∼N(µ 2 σ2
˙ X̄ = µ, σX̄ = n ) and
▶ T0 ∼N(µ
˙ 2 2
T0 = nµ, σT0 = nσ ).

12 / 21
5.4: The distribution of the sample mean
Example 5.27
The amount of a particular impurity in a batch of a certain chemical product is a
random variable with mean value 4.0 g and standard deviation 1.5 g. Suppose 50
batches are independently prepared.
▶ What is the approximate probability that the average amount of impurity is
between 3.5 and 3.8 g?

13 / 21
5.4: The distribution of the sample mean
Example 5.27 continued
The amount of a particular impurity in a batch of a certain chemical product is a
random variable with mean value 4.0 g and standard deviation 1.5 g. Suppose 100
batches are independently prepared.
▶ What is the approximate probability that the total amount of impurity in these
batches is at most 425 g?

14 / 21
5.4: The distribution of the sample mean
Example 5.29
▶ Distribution is non-normal with big spikes.
▶ Draw 1000 samples of n = 15 and calculate x̄ for each → Distribution is
approximately normal even for n < 30.

15 / 21
5.4: The distribution of the sample mean
Example 5.29 continued

16 / 21
5.4: The distribution of the sample mean
Other applications of the CLT: Binomial
▶ CLT can be used to justify the normal approximation to the binomial distribution.
▶ Let X ∼ Bin(n,
( p).
1 if the first trial results in S
Define X1 = , with X2 , ..., Xn defined
0 if the first trial results in F
similarly.
▶ Trials are independent and p is constant → Xi ’s are iid (and come from which
distribution?)
▶ Both X and Xn are approximately normally distributed when n is large (WHY?)
▶ Use this approximation only if n.p ≥ 10 and n(1 − p) ≥ 10.
▶ Whenever we are allowed to apply an approximately normal distribution, we will
have N(np, np(1 − p)) as our approximate distribution.

17 / 21
5.4: The distribution of the sample mean

Other applications of the CLT: Poisson

▶ Let Xi ∼ Poi( µn ), i = 1, 2, ..., n.
▶ It can be shown that X = X1 + X2 + ... + Xn ∼ Poi(µ) → CLT implies that a
Poisson rv with a sufficiently large µ has approximately a normal distribution
(Rule of thumb: µ > 20).
▶ Recall that for a Poisson distribution we have µ as BOTH the mean AND the
variance. Thus whenever we apply to an approximately normal distribution, we
will have N(µ, µ) as our approximate distribution.

Other applications of the CLT: Lognormal— Not applicable this semester

18 / 21
5.4: The distribution of the sample mean

Exercise 55, p.237

The number of parking tickets issued in a certain city on any given weekday follows a
Poisson distribution with parameter µ = 50.
▶ Calculate the approximate probability that between 35 and 70 (endpoints
included) tickets are given out on a particular day.

19 / 21
5.4: The distribution of the sample mean

Exercise 55, p.237 continued

The number of parking tickets issued in a certain city on any given weekday follows a
Poisson distribution with parameter µ = 50.
▶ Calculate the approximate probability that the total number of tickets given out
during a 5-day week is between 225 and 275 (endpoints included).

20 / 21
5.4: The distribution of the sample mean

Homework problems:
1. Let Xi ∼ Poi( µn ), i = 1, 2, ..., n. Show that E (X ) = µ if X = X1 + X2 + ... + Xn .
2. Exercise 48 (p.237, Probability and Statistics for Engineering and the Sciences,
Devore)
3. Exercise 56 (pp.237-238, Probability and Statistics for Engineering and the
Sciences, Devore)

21 / 21