Scribd
Upload
207 views

Proportional Stratified Random Sampling

A proportional stratified random sample of 1,000 units was taken from a population of 10,000 units divided into 4 strata. The sample was used to estimate the population mean of 15.3, the population total of 153,000, and the bound on error of estimation of 0.115139 and 11.5139 respectively. Variances were estimated for each stratum and combined using stratum sizes and sampling fractions to calculate total variances.

Uploaded by

manoharipl
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
207 views

Proportional Stratified Random Sampling

A proportional stratified random sample of 1,000 units was taken from a population of 10,000 units divided into 4 strata. The sample was used to estimate the population mean of 15.3, the population total of 153,000, and the bound on error of estimation of 0.115139 and 11.5139 respectively. Variances were estimated for each stratum and combined using stratum sizes and sampling fractions to calculate total variances.

Uploaded by

manoharipl
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 2

Proportional stratified random sampling, an example

How to estimate the population mean, the population total and


the bound, B, on the error of estimation for a population with
10,000 units divided into 4 strata.

The estimates are based on a sample of 1.000 units that was


proportional allocated to the strata.

Stratum Ni ni Ni /N Yi Ni /N* Y i
1 5,000 500 0.5 10.5 5.25
2 2,000 200 0.2 19.5 3.90
3 2,000 200 0.2 20.1 4.02
4 1,000 100 0.1 21.3 2.13
 10,000 1,000 1 15.3

The estimate of the population mean is

N1 N L
N
Y st  Y 1  ........  L YL   i Y i =15.3.
N N i 1 N

Stratum S2i S2i/ni (Ni -ni)/ Ni (Ni /N)2  N i   N i  ni  s i


2 2

N  N n
   i  i
1 16 0.032 0.9 0.25 0.0072
2 13.5 0.0675 0.9 0.04 0.00243
3 15 0.075 0.9 0.04 0.0027
4 10.3 0.103 0.9 0.01 0.000927
 0.013257

Estimated variance of Y st is

2
 N   N  ni  s i
 L 2
V (Y st )    i   i  = 0.013257
i 1  N   N i  ni
the Bound, B, on the error of estimation is

B=0.013257 = 0.115139

The estimated population total is

L
N Y st   N i Y i =10,000*15.3 =153,000
i 1

The estimated variance of N Y st :

  L
 N i  ni  si2
V ( N Y st )  N V (Y st )   N 
2
i
2
 = 10,000*0.013257 = 132.57
i 1  N i  ni

The Bound, B, on the error of estimation is:

B = 132.57 = 11.5139

You might also like