ESTIMATION OF PARAMETERS

Estimation is a kind of inference where we estimate the value of a population based on the information
obtained by the sample.

Note:
The parameter we will be estimating is only POPULATION MEAN (𝝁).

POINT ESTIMATE
The sample mean is the best point estimate of the population mean.
Example 1:
A researcher wants to estimate the average grade of all mathematics students in a certain school. He
determined the grades of five students as follows: 76, 82, 88, 90, and 96. Estimate the average
mathematics grade of all the students.
Solution:
Since the sample mean is the best point estimate for the population mean, we must compute for 𝑥 (sample
mean).
76 + 82 + 88 + 90 + 96
𝑥= = 86.4
5
Answer:
The estimated average mathematics grade of all students is 86.4.

Example 2:
Dominic wants to know how all the people in a district of Manila perceive the conduct of an earthquake
drill. He has interviewed 10 random selections from 500 people, and then, determined the percentages of
those who are in favor of the drill. He has assumed that the only error present is the sampling error. The
following numbers represent the mean percentages of the ten surveys.
90.0 86.4 90.1 85.2 78.0 95.3 90.0 87.5 93.0 77.5
What is his estimate of 𝜇?
1
Prepared by: Ms. Mary Abigail J. Aniban
Solution:
90.0 + 86.4 + 90.1 + 85.2 + 78.0 + 95.3 + 90.0 + 87.5 + 93.0 + 77.5
𝑥= = 87.3
10
Answer:
87.3 is his estimated mean percentage of the survey.

INTERVAL ESTIMATE
❑ Unlike point estimate, interval estimate is a range of values that contains the true population
value.
❑ There is a degree of confidence (usually a percent) that can be assigned before an interval estimate
is made. It is called CONFIDENCE LEVEL. It is the probability that the interval estimate will
contain the parameter assuming that a large number of samples is selected, and the estimation
process is done repeatedly.
In short, confidence level answers the question “How sure are we that the true value of the
population mean is within the interval estimate?”
Three commonly used confidence level: (90%, 95%, 99%)
❑ Interval estimate is also called CONFIDENCE INTERVAL.
If confidence interval is graphed on a normal curve, it will look like this:
90% confidence interval

𝜶 “alpha”: AREA OUTSIDE THE CONFIDENCE LEVEL

Note: Since we have two areas that are not part of the
confidence interval (left and the right tails), we divided 𝛼
by 2.

Critical value is the value that indicates the point beyond

which the true population parameter is not found. The
critical values can be a z-score or a t-score, depending on
the condition.

Note: Confidence interval is always considered as a two-

tailed curve.

𝜶
So, what is the 𝜶 and 𝟐 for the two remaining commonly used confidence level?
𝛼
95%: 𝛼 = 5% 𝑜𝑟 0.05 2
= 2.5% 𝑜𝑟 0.025
𝛼
99%: 𝛼 = 1% 𝑜𝑟 0.01 2
= 0.5% 𝑜𝑟 0.005

2
Prepared by: Ms. Mary Abigail J. Aniban
CONSTRUCTING A CONFIDENCE INTERVAL
When constructing a confidence interval, we have two conditions

▪ the population standard deviation (𝜎) is KNOWN

▪ the population standard deviation (𝜎) is UNKNOWN

Follow this diagram when constructing a confidence interval:

Example 3:
An electrical firm manufactures light bulbs that have life spans that are normally distributed, with a
standard deviation of 40 hours. If a random sample of 25 bulbs has a mean life span of 780 hours, find a
95% confidence interval for the population mean of all bulbs produced by this firm.
Solution:
First thing that we will determine is the population standard deviation. Is it known or unknown?
Analyze the sentence “An electrical firm manufactures light bulbs that have life spans that are normally
distributed, with a standard deviation of 40 hours.”. In this sentence, we talk about the life span of ALL
manufactured light bulbs and the standard deviation of their life span is found to be 40 hours. Since we
are pertaining to ALL light bulbs here, then the standard deviation describes the population. Hence, 40
hours is identified as POPULATION STANDARD DEVIATION.
Going back to the question, “IS THE POPULATION STANDARD DEVIATION KNOWN?” --- YES!

3
Prepared by: Ms. Mary Abigail J. Aniban
 𝜎 𝜎
If we check the given diagram, we have to use the formula (𝑋 − 𝑍𝛼⁄ , 𝑋 + 𝑍𝛼⁄ ).
2 √𝑛 2 √𝑛

Don’t get too overwhelmed with the formula. You can do this! To make our life easier, let’s list down the
given values.
Given:
𝜎 = 40 ℎ𝑜𝑢𝑟𝑠
𝑛 = 25 𝑙𝑖𝑔ℎ𝑡 𝑏𝑢𝑙𝑏𝑠
𝑥 = 780 ℎ𝑜𝑢𝑟𝑠
CL = 95% (confidence level)
𝛼 = 5% 𝑜𝑟 0.05
Recall: The critical values can either be a z-score or a t-score. Looking at the formula that we will be using
𝜎 𝜎
(𝑋 − 𝑍𝛼⁄2 𝑛 , 𝑋 + 𝑍𝛼⁄2 𝑛), we know for a fact that z-score will be our critical values.
√ √

So how do we get the critical values 𝒁𝜶⁄ ? For easier way of locating the critical values, look at the
𝟐
bottom part of your z-table.

Since the 𝜶 when we have 95% confidence level is 5% or 0.05, we will refer to the row under the two-
tailed column. Is it always two-tailed for confidence level? YES.
Therefore, our critical values are 𝑍𝛼⁄2 = ±1.96.

4
Prepared by: Ms. Mary Abigail J. Aniban
Now, we are ready to construct the interval estimate:
𝜎 = 40 ℎ𝑜𝑢𝑟𝑠
𝑛 = 25 𝑙𝑖𝑔ℎ𝑡 𝑏𝑢𝑙𝑏𝑠
𝑥 = 780 ℎ𝑜𝑢𝑟𝑠
𝑍𝛼⁄2 = ±1.96

𝜎 𝜎
(𝑋 − 𝑍𝛼⁄2 , 𝑋 + 𝑍𝛼⁄2 ) *substitute the values to this formula*
√𝑛 √𝑛

40 40
(780 − 1.96 , 780 + 1.96 ) = (764.32 , 795.68)
√25 √25

Note: 764.32 is the lower boundary and 795.68 is the upper boundary of the confidence interval.

Answer:
Therefore, the 95% confidence interval for the population mean of all bulbs produced by this firm is from
764.32 hours to 795.68 hours.

Example 4:
A random sample of 20 automobile owners shows that an automobile is driven, on an average, 23500
kilometers per year in the State of Virginia, with a standard deviation of 3900 km. Construct a 99%
confidence interval for the average number of kilometers an automobile is driven annually in Virginia.

Solution:
First thing that we will determine is the population standard deviation. Is it known or unknown?
Since the standard deviation of 3900 km is based on the random samples of 20 automobiles, then the
population standard deviation is UNKNOWN. What do we have then? Sample standard deviation
represented by symbol s.

5
Prepared by: Ms. Mary Abigail J. Aniban
 𝑠 𝑠
If we check the given diagram, we have to use the formula (𝑋 − 𝑡𝛼⁄ , 𝑋 + 𝑡𝛼⁄ ).
2 √𝑛 2 √𝑛

This is
chosen since
the sample
size is less
than 30 and
the
population
standard dev
is unknown.

Now, let’s list down the given.

𝑠 = 3900 𝑘𝑚
𝑛 = 20 𝑎𝑢𝑡𝑜𝑚𝑜𝑏𝑖𝑙𝑒
𝑥 = 23 500 𝑘𝑚
CL = 99% (confidence level)
𝛼 = 1% 𝑜𝑟 0.01 (two tailed)

Recall: The critical values can either be a z-score or a t-score. Looking at the formula that we will be using
𝑠 𝑠
(𝑋 − 𝑡𝛼⁄2 𝑛 , 𝑋 + 𝑡𝛼⁄2 𝑛), we know for a fact that t-score will be our critical values.
√ √

Quick Lesson about T-distribution!

T-distribution is used when the population standard deviation is unknown, and the sample size is
small (less than 30). It was formulated by W.S. Gosset and is sometimes called Student’s t
distribution

To get the t-score, the degrees of freedom, denoted by df, must be computed first. Degrees of
freedom are the number of values that are free to vary after a sample statistic has been computed.
The formula for the degrees of freedom is df = n-1, where n is the sample size.

So how do we get 𝒕𝜶⁄ ? Compute for the degrees of freedom.

𝟐

𝑑𝑓 = 𝑛 − 1 = 20 − 1 = 19
𝑑𝑓 = 19

6
Prepared by: Ms. Mary Abigail J. Aniban
Now, refer to the t-table.

Since the 𝜶 when we have 99% confidence level is 1% or 0.01, we will refer to the row under the two-
tailed column and the degrees of freedom of 19. Is it always two-tailed for confidence level? YES.
Therefore, our critical values are 𝑡𝛼⁄ = ±2.861.
2

Now, we are ready to construct the interval estimate:

𝑠 = 3900 𝑘𝑚
𝑛 = 20 𝑎𝑢𝑡𝑜𝑚𝑜𝑏𝑖𝑙𝑒
𝑥 = 23 500 𝑘𝑚
CL = 99% (confidence level)
𝑡𝛼⁄2 = ±2.861
𝑠 𝑠
(𝑋 − 𝑡𝛼⁄ , 𝑋 + 𝑡𝛼⁄ ) *substitute the values to this formula*
2 √𝑛 2 √𝑛

3900 3900
(23500 − 2.861 , 23500 + 2.861 ) = (21005.02 , 25994.98)
√20 √20

Answer:
The 99% confidence interval for the average number of kilometers an automobile is driven annually in
Virginia is from 21,005.02 km to 25,994.98 km.

7
Prepared by: Ms. Mary Abigail J. Aniban
Margin of Error
It is denoted by E. It is defined as the maximum difference between the observed sample mean and the true
value of the population mean.
𝜎 𝑠
In general, the term 𝑍𝛼⁄ and 𝑡𝛼⁄ describes the margin of error.
2 √𝑛 2 √𝑛

Try these!

1. Kim wants to estimate the average age of teachers in a certain town with the standard
deviation of 4. If she surveys a sample of 100 teachers and determined a sample mean of
35, compute a 90% confidence interval for the average age of teachers.
2. A recent survey conducted by a telecommunications company aims to determine the
average length of call (in minutes) per month. The researcher in charge surveyed 25
Filipinos at random and found that the average length of call they consume per month is
62 with a standard deviation of 4. Estimate a 90% confidence interval of the average length
of call (in minutes) per month.

8
Prepared by: Ms. Mary Abigail J. Aniban