Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Theme 5: Sampling Distributions

Statistics
Católica Porto Business School

1 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Basic Definitions

Parameter = a number that characterizes a population

(example: population mean µ) - it is typically unknown.

Statistic = a number that characterizes a sample.

(example: sample mean X̄ ) - we can calculate it from our
sample data.
We will use the information contained in the sample statistic
to make an inference about the population parameter.

2 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The Concept of a Sampling Distribution

Suppose that we want to estimate population mean µ

We can use the sample mean X̄ and the sample median M to

estimate µ

Which of these do you think will provide a better estimate of

µ?

3 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The Concept of a Sampling Distribution

Example: Toss a fair die. Let X denote the number of dots

showing on the up face. Suppose the die is tossed three times.
E(X)=µ=3.5
Sample 1: 2,2,6
X̄ =3.33
M=2
Sample 2: 3,4,6
X̄ =4.33
M=4
Neither the sample mean nor the sample median will always
fall closer to the population mean. So, we can not compare
these two sample statistics on the basis of their performance
with a single sample.
4 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The Concept of a Sampling Distribution

Since sample statistics are random variables, they must be

compared on the basis of their probability distributions.

What if we took all possible samples of the same size from the
same population, and each time, calculated the sample mean?
What would that set of X̄ values look like?

The probability distribution of a sample statistic is called the

sampling distribution of the sample statistic.

5 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The Concept of a Sampling Distribution

Example: In a single toss of a fair coin, let X denote the

number of heads observed. Consider a sample of 2 tosses (i.e,
the sample size is 2). Find the sampling distribution of X̄

Outcome Probability X̄

HH (X=1,X=1)

HT (X=1,X=0)

TH (X=0,X=1)

TT (X=0,X=0)

6 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Outcome Probability X̄

HH (X=1,X=1) 1/4 1

HT (X=1,X=0) 1/4 0.5

TH (X=0,X=1) 1/4 0.5

TT (X=0,X=0) 1/4 0

The sampling distribution of X̄ :

7 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The Concept of a Sampling Distribution

Populations often consist of a large number of different

values, making samples difficult (or impossible) to enumerate.
In this case, we might choose to obtain the approximate
sampling distribution for a sample statistic.
Example: The rolling machine of a steel manufacturer
produces sheets of steel varying in thickness. The thickness of
a steel sheet follows a uniform distribution with values
between 150 and 200 milimeters. Suppose we perform the
following experiment over and over again: Randomly sample
11 steel sheets from the production line and record the
thickness of each.
X1 +X2 +...+X10 +X11
Sample mean= X̄ = 11

Sample median=Sixth sample measurement when the 11

observations are arranged in ascending order

8 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The first 10 of the 1,000 computer generated samples from

the uniform distribution are presented in Table 6.4.

9 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

10 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The Properties of Sampling Distributions: Unbiasedness and Minimum Variance

The simplest type of statistic used to make inferences about a

population parameter is a point estimator.
Point Estimator: is a rule or formula that tells us how to use
the sample data to calculate a single number that can be used
as an estimate of the population parameter.
Example: the sample mean X̄ is a point estimator of the
population mean µ.
The sampling distribution of a point estimator provides
information about the point estimator.

11 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Unbiasedness

Suppose two point estimators, A and B, exist to estimate the

same population parameter θ. The sampling distributions are
given below:

Which sample statistic (A or B) is more attractive as an

estimator of θ?

12 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Unbiasedness

An estimator is called unbiased if the mean of its sampling

distribution is equal to the parameter being estimated.
If E (θ̂) = θ, then θ̂ is an unbiased estimator.
If E (θ̂) ̸= θ, then θ̂ is a biased estimator.
Biased estimators tend to either overestimate or
underestimate a parameter.
The sample mean X̄ is an unbiased point estimator for µ.
E (X̄ ) =

13 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Minimum Variance

Another nice property of an estimator: we want the spread of

its sampling distribution to be as small as possible.
The standard deviation of an estimator’s sampling distribution
is called the standard error of the estimator.
The standard error of the sample mean X̄ is √σ .
n

As the sample size gets larger, the spread of the sampling

distribution of X̄ gets smaller.

14 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Minimum Variance

15 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Point Estimator

Evaluating an estimator:
1 Is it unbiased?

2 Does it have a small standard error?

We desire an estimator that is unbiased and has the smallest
variance among all unbiased estimators. We call this statistic
the minimum-variance unbiased estimator (MVUE).

16 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The Sampling Distribution of X̄

We have determined the center and the spread of the

sampling distribution of X̄ .

µX̄ =E(X̄ ) = µ

σX̄ = √σn
What is the shape of its sampling distribution?
Case I: If the distribution of the original data is normal, the
sampling distribution of X̄ is normal.
This is true no matter what the sample size (n) is.

17 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Central Limit Theorem

Case II: Central Limit Theorem: If we take a random

sample (of size n) from any population with mean µ and
standard deviation σ, when the sample size is large, the
sampling distribution of X̄ is approximately normal with

mean: µX̄ = µ and

standard deviation: σX̄ = √σn

How large does n have to be?
Our rule of thumb: If n≥30, we can apply the Central Limit
Theorem.

18 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Central Limit Theorem

19 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Central Limit Theorem (CLT)

Why is the CLT important?

Because when X̄ is (approximately) normally distributed, we
can answer probability questions about the sample mean.
Standardizing values of X̄ :
√
If X̄ is normal with mean µ and standard deviation σ/ n,
then

X̄ − µ
Z= √
σ/ n
Z has a standard normal distribution.

20 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Example: A manufacturer of automobile batteries claims that the

distribution of the lengths of life of its best battery has a mean of
54 months and a standard deviation of 6 months. Suppose a
consumer group decides to check the claim by purchasing a sample
of 50 of the batteries and subjecting them to tests that estimate
the battery’s life.
Assuming that the manufacturer’s claim is true, describe the
sampling distribution of the mean lifetime of a sample of 50
batteries.

21 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Example: A manufacturer of automobile batteries claims that the

distribution of the lengths of life of its best battery has a mean of
54 months and a standard deviation of 6 months. Suppose a
consumer group decides to check the claim by purchasing a sample
of 50 of the batteries and subjecting them to tests that estimate
the battery’s life.
Assuming that the manufacturer’s claim is true, what is the
probability that the consumer group’s sample has a mean life
of 52 or fewer months?

22 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The Sampling Distribution of the Sample Proportion

Suppose that we want to estimate the proportion of voters in

favor of a bill to legalize gambling in a certain city or the
percentage of store customers who use credit cards to make a
purchase.

The data of interest are categorical in nature with two

outcomes (e.g., favor or do not favor).

Estimating proportion is equivalent to estimating the binomial

probability p.

23 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The Sampling Distribution of the Sample Proportion

The sample proportion, which is denoted by p̂, is a good

estimator of the population proportion p.

The point estimate of p is the sample proportion.

X
p̂ =
n
where X is the number of successes in the sample.

24 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The Sampling Distribution of the Sample Proportion

Suppose that we want to estimate the proportion of voters in

favor of a bill to legalize gambling in a certain city
Success: In favor of the bill→ 1
Failure: Not in favor of the bill→ 0
In a random sample of 100 people: 30 people are in favor of
the bill and 70 people are not in favor of the bill.
The sample proportion can be written as an average of 1s and
0s:

X 1 ∗ 30 + 0 ∗ 70
p̂ = = = 0.30
n 100
p̂ is the sample mean.

25 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The Sampling Distribution of the Sample Proportion

The sampling distribution of the sample proportion p̂ has

properties similar to those of the sampling distribution of X̄ .

E (p̂) =

V (p̂) =

The sampling distribution of p̂ is approximately normal when

the sample size is large (i.e., the Central Limit Theorem can
be applied when the sample size is large.)

26 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

The Sampling Distribution of the Sample Proportion

Mean of the sampling distribution:

E (p̂) = p

Standard deviation of the sampling distribution:

r
p(1 − p)
σp̂ =
n
For large samples, the sampling distribution is approximately
normal. A sample is considered large if

np ≥ 15 and n(1 − p) ≥ 15

27 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Example: According to a recent survey, 67% of adults who use

the Internet have paid to download music. In a random sample of
n=1000 adults who use the Internet, let p̂ represent the proportion
who have paid to download music.
Find the mean of the sampling distribution of p̂
Find the standard deviation of the sampling distribution of p̂

28 / 29
Basic Definitions The Properties of Sampling Distributions: Unbiasedness and Minimum Variance The Sampling Distribution of th

Example: According to a recent survey, 67% of adults who use

the Internet have paid to download music. In a random sample of
n=1000 adults who use the Internet, let p̂ represent the proportion
who have paid to download music.
What does the Central Limit Theorem say about the shape of
the sampling distribution of p̂?
Compute the probability that p̂ is greater than 0.5

29 / 29