Cairo University Special Program

Faculty of Computers and Artificial Intelligence Section#4

Probability and Statistics – 2 2022-2023

Sampling Distribution as Normal Distribution:

When a sampling unit is drawn from a finite population and is returned to that population, after
its characteristic(s) have been recorded, before the next unit is drawn, the sampling is said to be
“with replacement”. In the contrary case the sampling is “without replacement-Default”.

- If a population is normal with mean μ and standard deviation σ, the sampling distribution
means will always be normally distributed

𝜎
o with 𝜇𝑥̅ =𝜇 and 𝜎𝑥̅ = n is the sample size
√𝑛

o When population is finite If the sample is large relative to the population size, (n
is greater than 5% of N) “is n/N>0.05” and Sampling is without replacement,
Finite Population Correction is used

▪ Finite Population correction is correction factor used to modify the standard

error (i.e. the standard deviation for all sample means)

𝜎 𝑁−𝑛
▪ 𝜎𝑥̅ = × √
√𝑛 𝑁−1

- If the Population is not Normal, apply the Central Limit Theorem,

o Sample means (with or without replacements) from the population will be

approximately normal as long as the sample size is large enough (n > 25) otherwise,
we will use other distributions

𝜎
o with 𝜇𝑥̅ =𝜇 and 𝜎𝑥̅ =
√𝑛

o When population is finite If the sample is large relative to the population size, (n
is greater than 5% of N) and Sampling is without replacement, Finite Population
Correction is used

▪ Finite Population correction is correction factor used to modify the standard

error (i.e. the standard deviation for all sample means)
Cairo University Special Program
Faculty of Computers and Artificial Intelligence Section#4
Probability and Statistics – 2 2022-2023

𝜎 𝑁−𝑛
▪ 𝜎𝑥̅ = × √
√𝑛 𝑁−1
Problem 1:

If random samples of size three are drawn without replacement from a normally distributed
population consisting of four numbers 4, 5, 5, 7. Find sample mean 𝑋̅ for each sample and
make sampling distribution of 𝑋̅ . Calculate the mean of this sampling distribution and
Compare your calculations with population parameter.

Solution:
𝑁! 4! 4∗3!
Number of samples = NCn = 4C3 =𝑛!(𝑁−𝑛)! = 3!(1)! = 3!(1)! = 4 samples without replacements

Ps: number of samples in case of with replacement is 𝑁 𝑛

Sample No. Sample Values Sample Mean 𝑋̅

1 4, 5, 5 14/3
2 4, 5, 7 16/3
3 4, 5, 7 16/3
4 5, 5, 7 17/3

The mean of the population are:

∑ 𝑥 21
𝜇= = = 5.25
𝑁 4

14 16 16 17
+ + + )
𝜇𝑥̅ = ( 3 3 3 3 ) = 21 = 5.25
4 4

Then

𝜇 = 𝜇𝑥̅ = 5.25

That's why we use 𝜇 instead of 𝜇𝑥̅ in the sampling distribution because they are equal.
Cairo University Special Program
Faculty of Computers and Artificial Intelligence Section#4
Probability and Statistics – 2 2022-2023

Sampling Distribution:

X 14/3 16/3 17/3

Fre 1 2 1
P(X) 0.25 0.5 0.25

∑(𝑋 − 𝜇)2 (4 − 5.25)2 + (5 − 5.25)2 + (5 − 5.25)2 + (7 − 5.25)2

𝜎=√ =√ = 1.0897247
𝑁 4

(14/3 − 5.25)2 + (16/3 − 5.25)2 + (16/3 − 5.25)2 + (17/3 − 5.25)2

𝜎𝑥̅ = √ = 0.363241
4

Now, to calculate σ(Samples) using σ(Pop. ) →

Will we add the Standard Error?

1. Population is small “4 finite.”

2. n/N>0.05 (3/4=0.75)

3. Without replacement

𝜎 𝑁−𝑛
𝜎𝑥̅ = × √
√𝑛 𝑁−1
Cairo University Special Program
Faculty of Computers and Artificial Intelligence Section#4
Probability and Statistics – 2 2022-2023

𝑁−𝑛
Hence, √ was used because Sample size is larger than 5% of population and sampling
𝑁−1
without replacement.

1.0897247 4−3 1.0897247 1

𝜎𝑥̅ = × √4−1 = × √3 = 0.36324→ Standard Error
√3 √3

We can see that Standard Deviation of Samples means equal to Standard Error calculated using
Population parameters.

Problem 2:

A light-bulb manufacturer claims that the lifespan of its light-bulbs that follows normal distribution
has a mean of 54 months and a st. deviation of 6 months.
A) What is the probability that lifespan of light-bulbs is less 64 months ?
B) Your consumer advocacy group tests 50 of them. Assuming the manufacturer's claims are true,
what is the probability that it finds a mean lifetime of less than 52 months?

Solution:

A) P(x< 64)

z=64-54/6=1.666666667 -→ approximate to get from the table 1.67

P (z< 1.67) = 0.9525

B)

Sampling Distribution will follow Normal Distribution as population is normally distributed.

μ = 54

n = 50 sample size

σ = 6/√50 = 0.84852
Cairo University Special Program
Faculty of Computers and Artificial Intelligence Section#4
Probability and Statistics – 2 2022-2023

We need to find: P ( < 52)

(x − μ) 52 − 54
Z= = = −2.35
σ 0.84852
n

P (Z < -2.35) = .00939

Problem 3:

A brand name breakfast cereal company produces 5000 boxes of cereal per day. Each box is
supposed to have 368 grams of cereal with an average dispersion of 15 grams. Set up the
information in terms of population distribution. If a sample of 300 boxes in a certain day is
selected without replacement, what is the probability that the sample mean is less than 366?
Solution:
- Population are not mentioned as Normal
- Sample Size > 25
- Following Central Limit Theorem, Sampling Distribution will follow Normal
Distribution.
- Population Size = 5000 and Sample size = 300, Then Sample represent 6% of Population,
Then, Finite Population Correction factor is used.

μ= 𝜇𝑥̅ = 368. σ = 15 n = 300 N=5000

σ 𝑁−𝑛
σ = ∗√
√𝑛 𝑁−1

15 5000 − 300
= ∗√ = 0.86602540378 ∗ 0.96963293962 = 0.83972675805
√300 4999

We need to find: P ( < 366)

Cairo University Special Program
Faculty of Computers and Artificial Intelligence Section#4
Probability and Statistics – 2 2022-2023

(x − μ) 366 − 368
Z= = = −2.38
σ N −n 0.83972675
*
n N −1

P (Z < -2.38) = 0.0087

Problem 4:
In the long run, annual real returns on common stocks following a normal distribution pattern
have varied with mean 9% and standard deviation 28%. You plan to retire in 45 years and you
are considering investing in stocks. What is the approximate probability (assuming market
conditions do not change in 45 years) that the mean annual return on your investment over the
next 45 years will exceed 15%

Solution:

Population follows Normal Distribution; Then Sample will follow Normal Distribution.

μ = 0.09

σ = 0.28

Previous Population Parameters are calculated on long run.

n = 45

We need to find: P ( > 0.15)

Z=1.44
Cairo University Special Program
Faculty of Computers and Artificial Intelligence Section#4
Probability and Statistics – 2 2022-2023

P(Z >1.44) = 1- P(z<1.44) = 1- 0.9251= 0.0749