Scribd
Upload
75 views

04 - Interval Estimation

This document discusses interval estimation and methods for making statistical inferences from samples. It covers types of estimators like point and interval estimation. It also discusses topics like confidence intervals, estimating differences between population parameters, and choosing an appropriate sample size. Some key points are: - Interval estimation provides a range of values rather than a single point estimate for a population parameter. - Confidence intervals give probabilities that the estimated parameter will fall within calculated bounds. - Sample size impacts the accuracy and precision of estimates, with larger samples providing narrower intervals.

Uploaded by

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

04 - Interval Estimation

This document discusses interval estimation and methods for making statistical inferences from samples. It covers types of estimators like point and interval estimation. It also discusses topics like confidence intervals, estimating differences between population parameters, and choosing an appropriate sample size. Some key points are: - Interval estimation provides a range of values rather than a single point estimate for a population parameter. - Confidence intervals give probabilities that the estimated parameter will fall within calculated bounds. - Sample size impacts the accuracy and precision of estimates, with larger samples providing narrower intervals.

Uploaded by

Jason
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 42

INTERVAL ESTIMATION MATH 60.

2 – Introduction to
Statistical Analysis
WHEN WE INFER
Predicting short-term/long-term rates
Stock market behavior
Testing new product quality versus industry standards
Estimating prices
METHODS OF MAKING INFERENCES

Estimation – estimating or predicting the value of the parameter

Hypothesis Testing – Making a decision about the value of the


parameter based on some preconceived idea of what its value might be
TYPES OF ESTIMATORS
A rule, usually expressed as a formula, that tells us how to calculate an
estimate based on info in the sample, is called an estimator.

In point estimation, a single number is calculated


based on information in the sample.

In interval estimation, two numbers are calculated


based on sample data, forming an interval where the
parameter’s value is expected to lie.
EXAMPLE
Suppose that you want to find out the average weight of all students in
Ateneo College. You are able to select thirty students using a simple random
sampling and weigh them. The mean weight of the sample of students is 132
lbs with standard deviation of 9 lbs.
o The sample mean of 132 lbs can be used as a point estimate of the
population mean.
o The interval (123 lbs, 141 lbs) (1 standard deviation away from the mean)
can be used of as an interval estimate of the location of the population mean.
POINT ESTIMATES
UNBIASED VS BIASED

An estimator of a parameter is unbiased if the mean of


its distribution is equal to the true value. Otherwise, the
estimator is said to be biased.
MARGIN OF ERROR

If an estimator is
unbiased, the difference
between the point
estimate and the true
value will be less than
1.96 standard
deviation/error.
ESTIMATING A POPULATION MEAN
ESTIMATING A POPULATION
PROPORTION
EXAMPLE
One of the most famous large fractures (cracks) in the earth’s crust is
the San Andreas fault in California. A geologist attempting to study
the movement of the earth’s crust at a particular location found many
fractures in the local rock structure. In an attempt to determine the
mean angle of the breaks, she sampled n=50 fractures and found the
sample mean and standard deviation to be 39.8° and 17.2°,
respectively. Estimate the mean angular direction of the fractures and
find the margin of error for your estimate.
EXAMPLE
A random sample of 250 children between the ages of 8 and 18
showed that 170 children had a TV in their bedroom and that 120 of
them had a video game player in their bedroom.

a) Estimate the proportion of all 8- to 18-year-olds who have a TV in


their bedroom, and calculate the margin of error for your estimate.
b) Estimate the proportion of all 8- to 18-year-olds who have a video
game player in their bedroom, and calculate the margin of error
for your estimate.
CONFIDENCE INTERVALS

A much better method for estimating a parameter would be to


construct an interval that contains the true parameter value
As before, the sampled data may provide values that may be far
from the parameter value
What we can do is set the probability that the constructed
intervals will contain the estimated parameter
CONFIDENCE INTERVALS
The confidence coefficient, denoted by 1 – α,
is the probability that a confidence interval will
contain the estimated parameter.

1–α
CONFIDENCE INTERVALS
The confidence coefficient is NOT the
probability that the interval contains the
population parameter.
The more precise interpretation is that if
you construct 𝑀 confidence intervals,
(1 – α) 100% of those 𝑀 intervals will
possibly capture the parameter value.
REMARKS
 
CONFIDENCE INTERVALS

 
REMARKS
 
Confidence Coefficient and z-values

1-a z-value
0.90 1.645
0.95 1.96
0.98 2.33
0.99 2.58
EXAMPLE 1
 
EXAMPLE 2
A random sample of 10 chocolate energy bars of a
certain brand has, on average, 230 calories with known
population standard deviation 15 calories. Construct a
99% confidence interval for the true mean calorie content
of this brand of energy bar. Assume that the distribution of
calories is approximately normal.
EXAMPLE 3
A sample survey is designed to estimate the proportion of sports
utility vehicles being driven in the state of California. A random
sample of 500 registrations are selected from a Department of
Motor Vehicles database, and 68 are classified as sports utility
vehicles. Use a 99% confidence interval to estimate the proportion
of sports utility vehicles in California. How can you estimate the
proportion of sports utility vehicles in California with a higher
degree of accuracy?
ESTIMATING THE DIFFERENCE
BETWEEN TWO POPULATION MEANS

Consider comparing the following:


❑ The average scores of ACET for students whose major is Computer Science
and Management Information Systems
❑ The average yields of rice using two different types of fertilizer
❑ The average speed of AMD Ryzen 7 and Intel Core i7 8th gen processor
ESTIMATING THE DIFFERENCE
BETWEEN TWO POPULATION MEANS
ESTIMATING THE DIFFERENCE
BETWEEN TWO POPULATION MEANS
Example 1
Example 2
ESTIMATING THE DIFFERENCE
BETWEEN TWO POPULATION
PROPORTIONS
Consider comparing the following:
❑ The proportion of defective Iphone X from two production lines
❑ The proportion of male and female voters who favor same sex marriage
❑ The proportion of unvaccinated and vaccinated infants who get critically ill
from measles
ESTIMATING THE DIFFERENCE
BETWEEN TWO POPULATION
PROPORTIONS
Example 1
Example 2
In a study of the relationship between birth order and college success, an
investigator found that 126 in a sample of 180
college graduates were firstborn or only children. In a sample of 100
non-graduates of comparable age and socioeconomic background,
the number of firstborn or only children was 54. Estimate the
difference between the proportions of firstborn or only children in
the two populations from which these samples were drawn. Use a
90% confidence interval and interpret your results.
ONE-SIDED CONFIDENCE BOUNDS

Consider the following:


❑ A company wants to know if the interest rates that they need to pay the
bank from a loan will not exceed 8%
❑ An investor wants to know if the Return on Investment of a certain stock
portfolio will not go below 5%
ONE-SIDED CONFIDENCE BOUNDS
Example 1
A random sampling of a small grocery’s revenue for n=36
months produced a sample mean of PhP 54,740 and a
standard deviation of PhP 7,640. The owner is worried that
there will be months when he can’t pay the monthly
overhead cost of PhP 48,000. Does he need to worry? Use
a 99% confidence interval.
Example 2
A corporation plans to issue some short-term notes and is hoping
that the interest it will have to pay will not exceed 11.5%. To
obtain some information about this problem,
the corporation marketed 40 notes, one through each of 40
brokerage firms. The mean and standard deviation for the 40
interest rates were 10.3% and .31%, respectively.
Since the corporation is interested in only an upper limit on the
interest rates, find a 95% upper confidence bound for the mean
interest rate that the corporation will
have to pay for the notes.
CHOOSING A SAMPLE SIZE

Another important question centers on the construction of


the sample itself

❏ The statistic is dependent on the sampled data itself


❏ The confidence coefficient dictates the associated 𝑧-value
❏ The population standard deviation is a fixed parameter
❏ Only the sample size can be adjusted
CHOOSING A SAMPLE SIZE

● For estimating the population mean

● For estimating the population proportion

where B is the desired accuracy.


EXAMPLE 1
A college registrar wants to estimate the arithmetic mean
quality point index (QPI) of all graduating seniors during the
past 10 years. The mean QPI is to be estimated within plus or
minus 0.05 of the population mean. The standard deviation is
estimated to be 0.279. Using a 99% level of confidence, how
many transcripts should the registrar study?
Solution to Example 1
EXAMPLE 2
Suppose that you wish to estimate the binomial
parameter p to within 0.05 with a 95% confidence
bound. How large should the sample size be to
achieve this?
EXAMPLE 2
Note: We use p=0.5 to get the best accuracy.

You might also like