Scribd
Upload
12 views34 pages

Stat LL Chapter 2

Uploaded by

amanuelbalewgize0119
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
12 views34 pages

Stat LL Chapter 2

Uploaded by

amanuelbalewgize0119
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 34

CHAPTER TWO

STATISTICAL ESTIMATION

Managerial Statistics by Nugusu Meskele(MBA)


Introduction

 Statistical inference is the process of using sample results to


draw conclusions about the characteristics of a population.
 Inferential statistics enables you to estimate unknown
population characteristics such as a population mean or a
population proportion and population standard deviations.
 Statistical inference has two objectives; to estimate the
unknown population parameter and to test the hypothesis
about the population characteristics based on the sample
information.
Introduction,,,

 Thefocus here in this chapter is estimating a


population mean and the population
proportion with reasonable degree of accuracy.
 Estimate in statistics is made when information
available is incomplete or imperfect based on
sound judgment or experiences and when the
samples are scientifically selected.
Terminologies
 Estimation: is the process of making judgment or opinion about
the population characteristics from the information obtained from
the scientifically selected sample.
 Estimate: is a specific value/opinion made on the bases of
sample information about population.
 Estimator: a rule/formula that tells us how to estimate a value
for a population parameter using sample data.
 It is the sample statistic used to make decision or opinion about
population parameter.
Parameters Estimator Standard error
/statistics

Population mean (μ)

Difference between two


population means (µ1-µ2)

Population proportion
(P)
Types of Estimates

 Two types of estimates are used to estimate population


parameters: point estimates and interval estimates.
1. A point estimate is a single value of sample statistic,
which is used to estimate an unknown population
parameter.
 The limitation of point estimate is that it does not give
any idea about the reliability or precision of the
method of estimation used.
CONT…
2. A confidence interval estimate is a range of
values, called an interval, constructed around the
point estimate.
 The confidence interval is constructed such that the
probability that the population parameter located
somewhere within the interval is known.
The interval estimate indicates the value in two ways:
 By the extent of its range
 By the probability of the true population
parameter lying within that range.
Criteria for Good Estimator

CONT…

CONT…
Efficiency
 The efficiency of estimator is measured in terms of the size of the standard
error of the statistic.
 Estimator is necessarily characterized by a certain amount of variability.
 As the standard error of the sample statistic get smaller, the
estimator becomes more efficient.
 If the sample is large the standard deviation is to be divided by large
denominator, which in turn decreases the standard error.
Sufficiency
 A sufficient statistic is the one that utilizes all the information a sample
contains about the parameter to be estimated.
Point estimates of population mean and
variance.

Point Estimates of Population Proportion

 Because point estimator cannot be expected to provide the exact value of the population parameter; an
interval estimate is often computed by adding and subtracting a value, called the margin of error,
to the point estimate.
 The general form of an interval estimate is as follows:
Confidence interval = Point estimate ± Margin of error
 Interval estimate consists of two values (lower limit and upper limit) between which the true population
value lies with some stated level of significance.
 The level of significance (α) is the probability of being wrong in asserting that the confidence limits (1
-α) include the value of the true parameter.
 Each interval is constructed with regard to a given confidence level.
 Confidence level (1-α) is the probability of being correct in asserting that the confidence limits (1 - α)
include the value of the parametric statistic.

CONT..

Example
A random of 100 students belonging to the college
was taken. It was found that the mean height of these
students was 168.75cm and the standard deviation
was 7.5cm.What should be the confidence intervals
for estimating the mean height of the entire
population of students at 99% confidence?
CONT…


 When the sample n<30 or small sample, the sampling distribution is no
longer normal.
 In such a case, student t-distribution is used. Both normal and t-distribution
are symmetrical, but t-distribution is flatter than normal distribution.
 There is different t-distribution for each sample size.
 There should be n-1 degree of freedom for specified n sample size.
 Degree of freedom is the level of freedom to choose the values.
 The t-distribution doesn’t give the chance that a particular population
parameter will be within a specified confidence interval.
 Instead, it shows the chance that the particular population parameter will
not be within our confidence interval
CONT…
Example
A firm has appointed a large number of dealers all over
the country to sell its bicycles. It is interested in
knowing the average sales per dealer. A random sample
of 25 dealers is selected for this purpose. The sample
mean is 50,000birr and the standard deviation is 20,000
birr. Construct an interval estimate with 95%
confidence.
CONT…


Interval Estimate of the Difference between Two
Population Means
CONT…
Example
A research team is interested in the difference between the
productivity of employees with and without motivation scheme. A
sample of 6 employees offered different incentives and yielded an
average daily production of 30units/day. Other sample of 9
employees was found to have a mean daily production of
23units/day. Assume that the two populations of values are
normally distributed with standard deviation equal to 2 and 3,
respectively; find the 95% confidence interval for
CONT…

Interpretation
We are 99% confident that the true difference between the two population
means is found between 0.523 and 0.655 (0.523< - < 0.655).
CONT…

CONT…

Finite population correction factor


If the population sample is finite or the small sample is used, we
need to make some adjustments in the way we compute the
standard error of the sample means and standard error of the
sample proportions.
 For a finite population, where the total number of objects is N
and the size of the sample is n, the following adjustment is made
to the standard errors of the sample means and the proportion.
 n=Z2 pq where, e=error. n= sample size
 e2
CONT…
Choosing Appropriate Sample Size
 How many items should be in the sample? The correct sample size depends
on three factors: the level of confidence desired, the maximum allowable
error and the standard deviation of the population.
 The first factor is level of confidence that the researcher wants to have in
the result of the study. Since covering the entire population is not always
possible, we are satisfied with less than 100% confidence. Three confidence
levels are usually used: 99%, 95%, and 90%.
 The second is the maximum allowable error (E).
It is the amount that is added and subtracted from/to the sample statistics to
determine the end points of the confidence intervals.
 It is the amount of error the researcher is willing to tolerate.
 Small allowable error will require a large sample.
CONT…
 Thirdly, the population standard deviation is the other factors in determining
the sample size. If the population is widely dispersed, a large sample is required.
Thus, finding the population standard deviation may be necessary.
Three techniques are suggested for this purpose.
 Using comparable study approach when there is an estimate of the dispersion.
 A range-based approximation: all the observation could be expected to be
within ±3 standard deviation of the mean of the normal distribution. So the
distance between the largest and the smallest values is 6δ. Thus, we can estimate
the standard deviation as one-six of the range.
 Conducting a pilot test study to compute the sample standard deviation.
CONT…

You might also like