0% found this document useful (0 votes)
916 views49 pages

Estimation of Parameters

Point estimation involves finding a single value from a sample to approximate an unknown population parameter. The sample mean is a common point estimate of the population mean. Interval estimation provides a range of values that is likely to contain the true population parameter based on a given confidence level. The margin of error and critical value are used to calculate confidence intervals when the population variance is known. The t-distribution is used for interval estimation when the population variance is unknown and is based on the sample size, mean, and standard deviation.

Uploaded by

sheyn same
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
0% found this document useful (0 votes)
916 views49 pages

Estimation of Parameters

Point estimation involves finding a single value from a sample to approximate an unknown population parameter. The sample mean is a common point estimate of the population mean. Interval estimation provides a range of values that is likely to contain the true population parameter based on a given confidence level. The margin of error and critical value are used to calculate confidence intervals when the population variance is known. The t-distribution is used for interval estimation when the population variance is unknown and is based on the sample size, mean, and standard deviation.

Uploaded by

sheyn same
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/ 49

ESTIMATION

OF
PARAMETERS
POINT AND INTERVAL
ESTIMATION
What is point estimation?
 Point estimation is the process of finding the single value,
called point estimate, from a random sample of the
population, to approximate a population parameter. The
sample mean ẍ is the point estimate of the population mean
µ and 𝒔𝟐 is the point estimate of population variance 𝝈𝟐 .
 A point estimator is needed to obtain a point estimate.
It is assumed that the population is normally distributed
when the sample mean is used as a point estimate of the
population mean. The point estimator of a sample mean ẍ.
EXAMPLE: The following are the
systolic blood pressure of all teachers
in a private high school.
110 120 130 120 120
130 112 125 120 130
120 120 130 110 120
140 115 125 130 115
145 120 123 140 130
110 140 140 120 120
160 120 135 130 125
130 130 140 120 110
125 150 125 110 125
115 120 130 120 130
Compute the population mean.
X X X X X
120 110 120 130 120
130 112 125 120 130
120 120 130 110 120
140 115 125 130 115
145 120 123 140 130
110 140 140 120 120
160 120 135 130 125
130 130 140 120 110
125 150 125 110 125
115 120 130 120 130

TOTAL: 1295 1237 1293 1230 1225

σ𝒙
𝝁= =
𝑵
𝟏𝟐𝟗𝟓+𝟏𝟐𝟑𝟕+𝟏𝟐𝟗𝟑+𝟏𝟐𝟑𝟎+𝟏𝟐𝟐𝟓 𝟔𝟐𝟖𝟎
= = 𝟏𝟐𝟓. 𝟔
𝟓𝟎 𝟓𝟎
Assume that the following systolic blood
pressure were randomly selected from the
50 observations in the above example.

120 112 120 130 120


140 115 130 110 130
145 140 123 140 115
110 150 140 120 130
130 120 130 120 130
x x x x x
120 112 120 130 120
140 115 130 110 130
145 140 123 140 115
110 150 140 120 130
130 120 130 120 130

TOTAL: 645 637 643 620 625

σ𝑿
ഥ=
𝑿
𝒏
𝟔𝟒𝟓 + 𝟔𝟑𝟕 + 𝟔𝟒𝟑 + 𝟔𝟐𝟎 + 𝟔𝟐𝟓 3170
= = = 𝟏𝟐𝟔. 𝟖
𝟐𝟓 25
The sample mean 𝑥ҧ = 126.8 is still
different from the population mean 𝜇 =
125.6
Perhaps, it is better to approximate the
population parameter by determining a
range of values within which the
population parameter by determining a
range of values within which the
population mean is most likely to be
located. This range of values is called
confidence interval.
What is confidence level?
 Confidence Level refers to the
probability that the confidence interval
contains the true population
parameter.
The confidence level = 𝟏 −∝ 𝟏𝟎𝟎%
Where: α= probability that the
confidence interval does not contain
the true population parameter.
What is critical value?
Critical value is the value that
indicates the point beyond which
lies the rejection region. This
region does not contain the true
population parameter.
INTERVAL ESTIMATE OF
POPULATION MEAN WITH
KNOWN VARIANCE
 Formula for interval estimate of population mean
when population variance is known and n≥30.
𝜎 𝜎 𝜎 𝜎
 𝑥ҧ = 𝑧∝ , 𝑥ҧ + 𝑧∝ or 𝑥ҧ − 𝑧∝ < 𝜇 < 𝑥ҧ + 𝑧∝
2 𝑛 2 𝑛 2 𝑛 2 𝑛
Where:
 𝑥ҧ = mean of a random sample of size n
 n = sample size
 𝜎 = population standard deviation = 𝜎 2

 𝑧∝ = z value at (1 − ) confidence level
2 2
ҧ is:
 For a specific value, ∝= 0.05, the distribution of 𝑥’s
 The confidence interval can be written as
(𝑥ҧ − 𝐸, 𝑥ҧ + 𝐸) or (𝑥ҧ − 𝐸 < 𝜇 < 𝑥ҧ + 𝐸)
 Where:
 µ= population mean
 E = margin of error
 𝑥ҧ − 𝐸 = the lower confidence limit
 𝑥ҧ + 𝐸 = the upper confidence limit
 To find the margin of error, use the following formula:
𝜎
 𝐸= 𝑧∝ .
2 𝑛
Example
The mean score of a random sample of 49 of
Grade 11 students who took the first periodic test
is calculated to be 78. The population variance is
known to be 0.16.
Find the 95% confidence interval for the mean of
the entire Grade 11 students.
Find the lower and upper confidence limits.
Solution
 Step 1. Find the value of 𝑧∝ .
2
 Given 𝑥ҧ = 78, 𝜎 2 = 0.16, 𝑛 = 49.
 Confidence level is 95%.
 1 −∝ 100% = 95%
 1 −∝ 1 = 0.95
 1 −∝ = 0.95
 ∝= 0.05

 = 0.025 ▬► is divided by 2 since the interest is
2
on either left or right
 Step 2: Find 𝜎, then find the margin of error.
 𝜎 = 𝜎2
 = 0.16
 = 0.4
𝜎
 𝐸= 𝑧𝜎
2 𝑛
0.4
 = 1.96( )
49
 = 0.112

 Step 3: Substitute the values of 𝑥ҧ and E in the
confidence interval 𝑥ҧ − 𝐸 < 𝜇 < 𝑥ҧ + 𝐸
 78 − 0.112 < 𝜇 < 78 + 0.112
 77.888 < 𝜇 < 78.112
 77.89 < 𝜇 < 78.11
INTERPRETATION:
 The researcher is 95% confident that the
sample mean 𝑥ҧ = 78 differs from the population
mean 𝜇 by no more than 0.112 or 0.11. Also, the
researcher is 95% confident that the population
mean 𝜇 is between 77.89 and 78.11 when the mean
of the sample is 78.

SAMPLE SIZE FOR
ESTIMATING POPULATION
MEAN 𝝁
 The size of the sample is important in estimating the population mean 𝜇. The
following formula can be used to determine the appropriate sample size.
 Sample Size for Estimating 𝜇:
𝑧∝ 𝜎
2 2
 𝑛=[ ]
𝐸
 Where: 𝑧∝= critical value based on the desired confidence level
2
 E = desired margin of error
 𝜎 = population standard deviation




Example
Find the minimum sample size
required to estimate an unknown
population mean 𝜇 using the following
given data.
Confidence level = 95%
Margin of error = 75
𝜎 = 250
Solution
 Step 1: Find 𝑧∝ .
2

 Confidence level is 95%.


 1 −∝ 100% = 95%
1 −∝ 1 = 0.95
∝= 0.05

 = 0.025
2
0.5 − 0.025 = 0.475
Hence, 𝑧∝ = 1.96.
2
𝑧∝ 𝜎
2 2
Step 2: 𝑛 = [ ]
𝐸
1.96(250) 2
 = [ ]
75
 = 42.68 𝑜𝑟 43
The minimum sample size required to
estimate an unknown population mean
𝜇 using the given data above 43.
THE t-DISTRIBUTION
What is t-distribution?
The t-distribution, just like the standard normal curve, is bel
shape and unimodal. It is symmetric about t=0. However, its
variance is greater than 1. This makes its wider and flatter in
the middle. It has more are in its tails than that of the
standard normal curve. Its shaped depends on the sample
size n. As the sample size n becomes larger, the t-
distribution gets closer to the standard normal distribution.
The t-distribution formula is:
ҧ
𝑥−𝜇
 𝑡= 𝑠
𝑛
 Where:
𝑥ҧ = sample mean
𝜇 = population mean
S = standard deviation of the sample mean
n = sample size


EXAMPLE: A student researcher wants to
determine whether the mean score in mathematics
of the 25 students in Grade 8 Section Newton is
significantly different from the school mean of 89.
The mean and the standard deviation of the scores
of the students in Section Newton are 95 and 15,
respectively. Assume a 95% confidence level.
 SOLUTION:
 Step 1: Find the degrees of freedom.
 𝑑𝑓 = 𝑛 − 1
 = 25 − 1
 = 24
 Step 2: Find the critical value. Use the Table of t
Critical Values.
 Confidence level is 95%.
 1 −∝ 100% = 95%
 1 −∝= 0.95
 ∝= 0.05

 = 0.025
2
 Look at 24 under the column headed df. Move to the right along
the row until reaching the column headed 0.05, area in two tails
or 0.025 for area in one tail. The critical values in 2.064.
 Step 3: Compute the test statistic t.
ҧ
𝑥−𝜇
 𝑡= 𝑠
𝑛
95−89
 = 15
25

 =2
 The computed value of t is equal to 2 which is smaller
than the tabular value of 2.064. The value of the test statistic or
computed t value does not fall in the critical region. Therefore, the
mean score of Grade 8 section Newton in Mathematics is the
same with mean score of all the students taking up Grade 8
Mathematics.
 B. Identifying Percentiles Using t-Distribution Table
 - Percentiles can be identified using the t-critical
Values Table.
 EXAMPLE: The graph of a distribution with df = 15 is shown
below.




−𝑡1 0 𝑡1
 If the shade area on the right is 0.05, what is the area to the left of
𝑡1 ?
 What does 𝑡1 represent?
 Find the value of 𝑡1 .
 SOLUTION:
 1 −∝ 100% = 𝜌
 1 − 0.05 100 = 𝜌
 0.95 100% = 𝜌
 𝜌 = 95%
 𝜌 = 0.95 area to the left of 𝑡1
 Hence 𝑡1 represents the 95th percentile 𝑡0.95 .
 To find the value of 𝑡1 , look at 15 under the column
headed df. Proceed to the right until the column
headed 0.05 for one tail is reached. The result is
1.753.
INTERVAL ESTIMATE OF POPULATION
MEAN WITH UNKNOWN VARIANCE
 If the population standard
deviation is unknown and the sample size
ҧ
𝑥−𝜇
is small, use the formula: 𝑡 = 𝑠 . This
𝑛
formula, when manipulated algebraically
will produce the following formula that is
used to determine the confidence interval
for estimating the population mean when
the population variance and population
standard deviation are unknown and the
sample size n is less than 30.
If n < 30 and 𝜎 is unknown, the
confidence interval for population
mean 𝜇 is:
𝑠 𝑠
(𝑥ҧ − 𝑡∝ , 𝑥ҧ + 𝑡∝ ) or 𝑥ҧ −
2 𝑛 2 𝑛
𝑠 𝑠
𝑡∝ < 𝜇 < 𝑥ҧ + 𝑡∝
2 𝑛 2 𝑛
Where: 𝑥ҧ = mean of a random
sample size n n= sample size
The confidence interval can also be
written as:
(𝑥ҧ − 𝐸, 𝑥ҧ + 𝐸) or 𝑥ҧ − 𝐸 < 𝜇 < 𝑥ҧ + 𝐸
Where: 𝑥ҧ − 𝐸 = lower confident limit
𝑥ҧ + 𝐸 = upper confident limit
𝜇 = population mean
𝐸 =margin error
To find the margin of error, use the
formula:
𝑠
𝐸 = 𝑡∝
2 𝑛
Where 𝑡∝ has n – 1 degrees of
2
freedom.
EXAMPLE: The mean and standard
deviation of the content of a sample of 10
similar containers are 10.5 liters and
0.352, respectively.
Find a 95% confident interval for the
actual mean content.
Find the lower and upper confident limits.
Assume that the contents are
approximately normally distributed.
 SOLUTION:
 Step 1: Find the degrees of freedom df.
 𝑑𝑓 = 𝑛 − 1
 = 10 − 1
 =9
 Step 2: Find ∝ in 1 −∝ 100% confidence level.
 1 −∝ 1 = 0.95
 1 −∝= 0.95
 ∝= 0.05

 = 0.025
2
 Step 3: Find the critical value of 𝑡∝ using the
2

Table of t-Critical Values.


The critical value is 𝑡∝ = 𝑡0.025 = 2.262
2
 Step 4: Find the margin of error E.
𝑠
 𝐸= 𝑡∝
2 𝑛
0.352
 = 2.262( )
𝑛
 = 0.2517 𝑜𝑟 0.25
 Step 5: Find the confidence interval.
 𝑥ҧ − 𝐸 < 𝜇 < 𝑥ҧ + 𝐸
 10.5 − 0.25 < 𝜇 < 10.5 + 0.25
 10.25 < 𝜇 < 10.75
 Step 6: Lower confidence limit = 10.25
 Upper confident limit = 10.75
POPULATION PROPORTION
A. What is sample proportion?
- The sample proportion 𝜌ො can be calculated as
long as the number of elements x in the sample
having the same characteristics and the size of
the sample are known. The sample proportion is
the point estimate of the population proportion 𝜌.
The population proportion 𝜌 can be viewed as
the proportion of the element in the population
possessing the characteristics of interest.
A point estimator of the population
proportion 𝜌 is given by the statistic:
𝑥
𝑝ො =
𝑛
Where: 𝑝ො = sample proportion
𝑥 = number of elements in the
sample having the same characteristics
𝑛 = sample size
EXAMPLE: In a random sample of
120 teachers, 48 of them have
master’s degrees. Find the value
of 𝜌.

SOLUTION:
𝑥

𝑃=
𝑛
48
=
120
= 0.40
The value of 𝜌ො which is 0.40 is the point
estimate of the population proportion 𝜌.
The formula for a confidence interval
for population proportion
ෝ (1−𝜌
𝜌 ෝ) ො
𝑝(1− ෠
𝑃)
𝜌ො − 𝑧∝ , 𝜌ො + 𝑧∝
2 𝑛 2 𝑛
Where: 𝑝ො = sample proportion
𝑛 = sample size
𝑧∝ = 𝑧 value corresponding to an area
2

in the right tail of a standard normal z
2
distribution.
The confidence interval can be written as
𝑝ො − 𝐸, 𝑝ො + 𝐸 𝑜𝑟 𝑝ො − 𝐸 < 𝑝 < 𝑝ො + 𝐸
Where: 𝑝 = population proportion
𝐸 = margin of error
𝑝ො − 𝐸 = lower confident limit
𝑝ො + 𝐸 = upper confident limit
To find the margin of error, use the
following formula

𝑝(1− ො
𝑝)
𝐸 = 𝑧∝
2 𝑛
EXAMPLE: Calculate the confidence
interval to estimate the population
proportion, given the following data.
𝑛 = 400 𝑝ො = 0.70
 SOLUTION:
 Step 1: Find ∝ in 1 −∝ 100% confidence level then find 𝑧∝.
2
 1 −∝ 100% = 95%
 1 −∝ = 0.95
 ∝= 0.05
∝ 0.05
 = = 0.025
2 2
 Subtract 0.025 from 0.500
 0.500 − 0.025 = 0.475
 Using the areas under the standard normal curve table 𝑧∝ = 1.96
2
 Step2: Find the margin of error 𝐸.

𝑝(1− ො
𝑝)
 𝐸 = 𝑧∝
2 𝑛

0.70(1−0.70)
 = 1.96
400

 = 0.045
 Step 3: Find the confidence interval.
 𝑝Ƹ − 𝐸 < 𝑝 < 𝑝Ƹ + 𝐸
 0.07 − 0.045 < 𝑝 < 0.7 + 0.045
 0.0655 < 𝑝 < 0.745
LENGTH OF CONFIDENCE INTERVAL
AND APPROPRIATE SAMPLE SIZE
The length of a confidence interval is the
absolute difference between the upper
confidence limit and the lower confidence limit.
That is,
𝐿𝐶𝐼 = 𝑈𝐶𝐿 − 𝐿𝐶𝐿
Where: 𝐿𝐶𝐼 = Length of Confident Interval
𝑈𝐶𝐿 = Upper Confident Limit
𝐿𝐶𝐿 = Lower Confidence Limit
EXAMPLE: Find the length of the
following confidence intervals
0.357 < 𝑝 < 0.603
SOLUTION: 𝑳𝑪𝑰 = 𝑼𝑪𝑳 − 𝑳𝑪𝑳
 = 0.603 − 0.357
 = 0.246
Determining the Sample Size for
Estimating 𝝆
When an estimate 𝒑
ෝ is known:
[𝒛∝ ]𝟐 𝒑
ෝ(𝟏−ෝ
𝒑)
𝟐
 𝒏=
𝑬𝟐
 Where: 𝒏 = desired sample size
 ෝ = sample proportion
𝒑
 𝒛∝ = z value
𝟐

 𝑬 = margin of error
 When an estimate 𝒑
ෝ is unknown
𝟎.𝟐𝟓(𝒛∝ )𝟐 (𝒛∝ )𝟐
𝟐 𝟐
 𝒏= =
𝑬𝟐 𝟒𝑬𝟐
 Where: 𝒏 = desired sample size
 ෝ = sample proportion
𝒑
 𝒛∝ = z value
𝟐

 𝑬 = margin of error
 0.025 is constant
 If the computed sample size is not a whole number,
it should be rounded up to the next whole number.
EXAMPLE: Find the sample size required to estimate
a population proportion given the following data
Confidence level = 90%
𝑝 = 0.48
𝐸 = 0.042
 SOLUTION:
 Step 1: Find ∝ in 1 −∝ 100% confidence level then find 𝑧∝ .
2

 1 −∝ 100% = 90%
 1 −∝= 0.90
 ∝= 0.10
∝ 0.10
 = = 0.05
2 2
 0.500 − 0.05 = 0.45
 Using the areas under the standard normal curve table 𝑧∝ = 1.645
2
[𝒛∝ ]𝟐 𝒑
ෝ(𝟏−ෝ
𝒑)
𝟐
 Step 2: 𝒏 =
𝑬𝟐
 =
𝟏.𝟔𝟒𝟓 𝟐 𝟎.𝟒𝟖(𝟏−𝟎.𝟒𝟖)
(𝟎.𝟎𝟒𝟐)𝟐
 =
𝟑𝟖𝟐. 𝟖𝟗𝟑𝟑 𝒐𝒓 𝟑𝟖𝟑

You might also like