Scribd
Upload
98 views

Sampling and Sampling Distribution of Sample Means

This document discusses sampling and estimation in statistics. It covers key concepts like population and sample, probability and non-probability sampling, simple random sampling, stratified random sampling, sampling distribution of the sample mean, and standard error of the sample mean. The document provides examples and formulas to illustrate these statistical concepts. It is intended for a university course on business and finance statistics.

Uploaded by

rakhshanda
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
98 views

Sampling and Sampling Distribution of Sample Means

This document discusses sampling and estimation in statistics. It covers key concepts like population and sample, probability and non-probability sampling, simple random sampling, stratified random sampling, sampling distribution of the sample mean, and standard error of the sample mean. The document provides examples and formulas to illustrate these statistical concepts. It is intended for a university course on business and finance statistics.

Uploaded by

rakhshanda
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

9/04/2016

STATISTICS FOR BUSINESS AND FINANCE


(BUS5SBF)

Topic 4: Chapter 6 (Pg. 247)


Time Value of Money
Sampling and Estimation

Topic 1:

Instance coordinator/Lecturer: Dr. Saqib Manzoor


E-mail: [email protected]

Subject coordinator: Associate Professor Ishaq Bhatti


E-Mail: [email protected]
La Trobe Business School
Slides have been drafted by the La Trobe University, La Trobe Business School based
on the Text book DeFusco et al (2015)
ICAEW 2013

Sampling

In statistics we are often interested in obtaining information about the value of


some parameters of a population (Size of pop. Is N).

To obtain this information we usually take a small subset of the population called
sample and try to draw some conclusions from this sample (Sample size is n).
A sampling plan is the set of rules used to select a sample. For instance,

i.

ii.

iii.

Simple Random Sampling

Stratified Random Sampling


Cluster Sampling

Sampling is a process of selecting a Sample from Population.

Therere two main methods of sampling (i) Probability Sampling

(ii) Non-Probability Sampling: Quota, personal judgement, no scientific justification,


just a personal choice sampling frame

9/04/2016

Probability Sampling
(1) Simple Random Sampling (SRS)
A simple random sample (n) is a subset of a larger population (N) created in
such a way that each element of the population has an equal probability
(1/N) of being selected in the sample.
Sampling With Replacement (WR);
- Elements will be repeated

- Number of samples : m N n

Example: Draw all possible samples of size 2 WR from Population = { A,B,C};


N =3, n =2 & Sampling With Replacement. Thus, m=9 samples
Samples: {AA,AB,AC,BB,BA,BC, CC,CA,CB}

Probability Sampling
(1) Simple Random Sampling (SRS)

Sampling Without Replacement (SWOR)

-Element once chosen will not be replaced in the population

Example: We need a president, vice president and a treasurer for a committee


comprising of 5 members. However, a person can not have two or more roles;
thus Sampling without replacement is required.
Number of possible samples:

Selecting 3 out of 5 members.

M C

N
N!

n
(
N

n )! n!

N = 5, n = 3 Sampling without replacement: M 3 (3!2!) 10


Let population be {A, B, C, D, E}

5!

Possible Samples: {ABC, ABD, ABE, ACD, ACE,ADE, BCD, BCE, BDE, CDE}

9/04/2016

2. Stratified Random Sampling

Stratified random sampling occurs when the population is


divided into subpopulations (strata) and a simple random sample
is drawn from each strata.
It guarantees that population subdivisions of interests are
represented in the sample.

It generates more accurate estimates (smaller variance) than


simple random sampling implies less volatility as compared to
SRSWR or SRSWOR

EXAMPLE (please take 10 samples from this population)

Simple random Sampling

Sample could have household 1 10, or


11- 20, or any other collection.

Stratum-I
Stratum-II
IMPORTANT ANNOUNCEMENT: CHECK YOUR SAMPLE
IN THE ASSIGNMENT FOR ANY REPETITIONS PLEASE

However, it would not represent the Male


and Female.
Stratified random sampling (based on
Gender of head , i.e., the population is
separated into mutually exclusive layers or
strata.)
Since we have equal proportions, our
sample with be stratified in the same
proportions.
We have to take 5 from no 1 -10 and
another 5 from no 11-20.

It would represent the proportion group


of male and female

9/04/2016

Types of Sample Data

Cross-sectional data represent observations over individual units


at a point in time
- For example, GDP of every country in year 2015

Time series data is a set of observations on a variables outcomes


in different time periods
- For example, GDP of Australia from 1960 to 2015

Panel data have both time-series and cross-sectional aspects and


consist of observations through time on a single characteristics of
multiple observational units.
- For example, GDP of every country from 1960 to 2015

Sampling Error and Statistic

Sampling error is the difference between the observed


value of a statistic and the quantity it is intended to
estimate. Sample size estimate: n to N
Sampling Error in case of pop mean;
X Sampling Error
Sample mean over estimates pop mean; ( X ) 0
Sample mean under estimates pop mean; ( X ) 0

Sampling distribution of a statistic is the distribution of all


the distinct possible values that the statistic can assume
when computed from samples of the same size; n randomly
drawn from the same population.

9/04/2016

Example: Distribution of Sample Mean


Suppose we have a 'population' of 5 elements with values 1, 2, 3, 4, 5
What are the average and standard deviation of this population?

X 3
X 1.41

Now, consider all possible samples of size 3 without replacement to provide a


point estimate of the population mean, and find the average of each sample.

Example: Distribution of Sample Mean

Now, consider all possible samples of size 3 to provide a point


estimate of the population mean, and find the average of each
sample. What is the average and standard deviation of the new
distribution?
Possible Samples, Size 3
1, 2, 3
1, 2, 4
1, 2, 5
1, 3, 4
1, 3, 5
1, 4, 5
2, 3, 4
2, 3, 5
2, 4, 5
3, 4, 5

2
2.33
2.67
2.67
3
3.33
3
3.33
3.67
4

SUM = 30

9/04/2016

Sampling Distribution of Sample Mean

( )

0.1

( )

0.2

( )

0.4

2.33

0.1

0.233

0.54289

0.2

0.6

1.8

2.67

0.2

0.534

1.42578

3.33

0.2

0.666

2.21778

0.1

0.4

1.6

3.67

SUM

10

0.1
1

Hmmm! Unbiased estimator

0.367
3

1.34689
9.3333

Sample Mean and Variance


E( X )

xf ( x ) 3
x

for all X

E( X 2 )

f ( x ) 9.3334

Var X E ( X 2 ) E ( X )
for all X

9.3333 9 0.3333
Verificati on

x E X 3

N n
N 1

1.414 5 3
0.3333
10
5 1
Hence verified .

Shape of the Sampling Distribution of Sample Mean

If n is large enough (>30) the sampling distribution of X will be


a normal distribution regardless of the distribution type
exhibited by the population.
If the population distribution is normal the sampling
distribution of X will be normal regardless of the sample size.

9/04/2016

Standard Error of the Sample Mean

When we use the sample mean to estimate the population


mean, there are some errors.
The standard error of the sample mean is the standard
deviation of the sample mean.

For a sample mean calculated from a sample generated from a


population with standard deviation , the standard error of the

sample mean is
; with replacement

SE X X

n
N n
; without replacement
N 1

when population standard deviation () is known.


What happens to SE X when n gets very large?

Standard Error of the Sample Mean

In practice, the population variance is almost always


unknown. The standard error of the sample mean is
estimated by,

sX s

Note,

n , where s 2

sX X

X
n

i 1

( n 1)

You might also like