0% found this document useful (0 votes)
718 views17 pages

03 Chapter 3 - Statistical Estimation

This chapter discusses statistical estimation and related concepts. There are two main types of estimation: point estimation and interval estimation. Point estimation provides a single value from a sample to estimate an unknown population parameter, while interval estimation provides a range of values within which the population parameter is likely to fall, given a certain confidence level. The chapter covers properties of good estimators like being unbiased. It presents formulas for constructing confidence intervals around point estimates of means, differences of means, and proportions. The final section discusses how to determine the needed sample size based on desired confidence level, population variability, and maximum tolerable error. Examples are provided to illustrate key statistical estimation techniques.

Uploaded by

Yohanna Sisay
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
718 views17 pages

03 Chapter 3 - Statistical Estimation

This chapter discusses statistical estimation and related concepts. There are two main types of estimation: point estimation and interval estimation. Point estimation provides a single value from a sample to estimate an unknown population parameter, while interval estimation provides a range of values within which the population parameter is likely to fall, given a certain confidence level. The chapter covers properties of good estimators like being unbiased. It presents formulas for constructing confidence intervals around point estimates of means, differences of means, and proportions. The final section discusses how to determine the needed sample size based on desired confidence level, population variability, and maximum tolerable error. Examples are provided to illustrate key statistical estimation techniques.

Uploaded by

Yohanna Sisay
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 17

Chapter 3: Statistical Estimation

3.1. Types of Statistical Estimation


❑Goal: How can we use sample data to estimate values
of population parameters?
❑ Two types of estimation:
1) Point estimate: A single value that computed from
a sample to estimate a population parameter
2) Interval estimate – is an interval of numbers
around the point estimate, that has a fixed
“confidence level” of containing the parameter
value, called a confidence interval.
Most common point estimators
Population Sample

Mean m mˆ = y =  y i

Standard deviation
s sˆ = s =  i
( y − y ) 2

n −1
3.2. Properties of Estimators
• Unbiased: Sampling distribution of the estimator
centers around the parameter value E ˆ  =  for all   

• A sufficient statistic, Y, is a statistic which contains all


the information for the estimation of .
• Efficient: Smallest possible standard error, compared
to other estimators
Example: Point estimate
• To set the price of a product, one strategy is a competition-
oriented in which you fix the price of your product at the
average level charged by other producers. Suppose you want to
market a 200 gram bar soap that you produce. The current
wholesale prices from a random sample of 10 soap producers
(Birr) are:
1.00 1.35 1.50 0.95 0.90
1.25 1.00 1.20 0.90 1.50
a) What is an estimate of the mean wholesale price charged by
all producers? (Ans. = 1.155)
b) Find an estimate of the standard deviation in the whole prices
of all producers. (Ans. = 0.238)
3.3. Confidence Interval for the Population Mean

• Confidence interval – is a range of values constructed from


sample data so that a parameter occurs within that range at a
preselected probability. The preselected probability is called
“level of confidence”.
• Note:
– 90% of the sample means lie within 1.65 standard
deviation of the population mean
– 95% of the sample means lie within 1.96 standard
deviation of the population mean
– 99% of the sample mean will lie within 2.58 standard
deviation of the population mean
Example: confidence interval for population mean
3.4. Confidence Interval for the Difference of Means

• If 𝒙
ഥ𝟏 and 𝒙
ഥ𝟐 are sample means based on large
and independent random samples of size 𝒏𝟏 ≥
𝟑𝟎and 𝒏𝟐 ≥ 𝟑𝟎from two infinite populations
with means 𝝁𝟏 and 𝝁𝟐 with standard deviation
𝝈𝟏 and 𝝈𝟐 , respectively, then with ሺ𝟏 −
𝜶ሻ𝟏𝟎𝟎%confidence interval for the difference of
population means (𝝁𝟏 - 𝝁𝟐 ) is given by ሺഥ
𝒙𝟏 -

𝝈𝟐 𝟏 𝝈𝟐 𝟐
ഥ 𝟐 ሻ ± 𝒁𝜶
𝒙 +
𝟐 𝒏𝟏 𝒏𝟐
Example: difference between two population means

• A sample of 150 brand A light bulbs showed a mean


life time of 1,400 hrs and a standard deviation of 120
hrs. A sample of 200 brand B light bulbs showed a
mean life time of 1,200 hrs and a standard deviations
of 80 hrs.
a) Find a 95% and 99% confidence interval for the
difference of the mean lifetimes of the populations
of brand A and brand B.
Ans. 95% confidence interval is (175.2, 224.8)
Ans. 99% confidence interval is (167.4, 232,6)
3.5. Confidence Interval for Proportions

• A 𝟏 − 𝜶 𝟏𝟎𝟎%confidence interval for the


𝑷ሺ𝟏−𝑷ሻ
population proportion 𝝅 is found by 𝑷 ± 𝒁𝜶 𝒏
𝟐
where P = sample proportion
𝒁𝜶 =the standard normal value corresponding to
𝟐
the described level of confidence
n = sample size (i.e., 𝒏 ≥ 𝟑𝟎ሻ
Example: confidence interval for population
proportion
• Suppose 1,600 of 2,000 registered voters sampled
said they planned to vote for the Prosperity Party
candidate for Mayor of Addis Ababa. Using 95%
degree of confidence, what is the interval estimate
for the population proportion?
➢ Ans. = (0.7825, 0.8175)
➢ Interpretation: if election were held today, we are
95% sure that the Prosperity Party candidate for
mayor of Addis Ababa would receive between
78.2% and 81.8% of the vote.
3.6. Sample Size Determination
❑ The correct sample size depends on three factors:
1) The level of confidence desired
2) The variability in the population being studied
3) The maximum allowable error
❑ The maximum allowable error (E)
✓ is the amount of error that the researcher is willing to
tolerate
✓ is the amount that was added and subtracted from the
sample mean to obtain the limits of the confidence
interval.
❖ A small allowable error will require a large sample size,
where as a large allowable error will permit smaller sample
sizes.
3.6.1. Sample Size Determination – estimating population mean

• Sample size determination when estimating


𝒁∗𝑺 𝟐
population mean will be 𝒏 =
𝑬
where,
E = the maximum allowable error
S = the estimate of the population standard deviation
n = sample size
Z = standard normal value corresponding to the desired
level of significance
Note:
❖ Because the result of this computation is NOTalways a whole
number, the usual conservative practice is to round up any
fractional result
Example: Sample Size Determination – estimating population mean
A study is to be conducted on the mean salary of mayors of
cities with a population of fewer than 100,000. The error (E) in
estimating the mean is to be less than Birr 100 and a
confidence level of 95% is desired. Suppose that the standard
deviation of the population is estimated to be Birr 1,000. What
is the required sample size?
Solution:
E = Birr 100
The value of Z for a 95% confidence = 1.96
S = Birr 1,000
𝒁∗𝑺 𝟐
The required sample size 𝒏 = = 384.16 = 385
𝑬
❖ If a higher level of confidence were desired, say 99%, then
larger sample would be required n = 665.64 = 666. Check!
3.6.2. Sample Size Determination – for proportions

• To determine the sample size for proportion, we use


𝒁 𝟐
the following formula 𝒏 = 𝑷ሺ𝟏 − 𝑷ሻ
𝑬
• The value of Pcan be estimated from pilot study. If no
estimate is available, the value of 0.50 is used. Why?
Because the term P(1-P) can never be larger than
when P = 0.50.
Example: We are planning a survey to find the proportion of
cities that have a private garbage collectors. We want a
maximum error (E) of 0.10 and 90% confidence level, and
tentatively estimate the proportion (P) at 0.5. What is the
required sample size? Take 90% confidence level as at 1.65.
𝟏.𝟔𝟓 𝟐
𝐧 = 𝟎. 𝟓ሺ𝟏 − 𝟎. 𝟓ሻ
𝟎.𝟏𝟎
= 68.0625 = 69
END OF CHAPTER 3

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy