Scribd
Upload
10 views20 pages

Lecture 9 - March 4

Uploaded by

adhithxt
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
10 views20 pages

Lecture 9 - March 4

Uploaded by

adhithxt
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/ 20

MA2540/MA4240: Applied Statistics

Dr. Sameen Naqvi


Department of Mathematics, IIT Hyderabad
Email id: [email protected]
Confidence Interval Estimation
Agenda

I Understand the concept of Confidence Intervals

I Find confidence interval for

I Population Mean

I Population Variance

I Population Proportion
Confidence Interval

I The Situation:
Point estimates, such as the sample proportion (p̂), sample
mean (x), and the sample variance (S 2 ), depend on the
particular sample selected.

I The Problem:

When we use x to estimate µ, can we be confident that x is


close to µ?

When we use p̂ to estimate p, can we be confident that p̂ is


close to p?
Confidence Interval

I A solution:
Rather than finding point estimate, we could find an interval
(or range) of values that we can be really confident contains
the actual unknown population parameter.

I So, we find lower (L) and upper (U) values between which we
can be really confident the population mean/proportion falls:

L<µ<U or L < p < U.


Confidence Interval

I An interval of such values is referred to as a confidence


interval.

I Each interval has a confidence coefficient (reported as a


proportion):
1−α
or a confidence level (reported as percentage):

(1 − α)100%

I Typically, confidence coefficients are 0.90, 0.95, and 0.99,


with corresponding confidence levels as 90%, 95%, and 99%.
Interpretation of Confidence Interval (CI)

Suppose you calculate CI for mean with 95% confidence level.

Then, we can be 95% confident that the population mean


falls between L and U.

It does not imply that there is a 95% probability that the


interval would contain the true population parameter.
Interpretation of Confidence Interval

I In real-world, we take just one random sample.

So, the interval that we obtain can be either correct or


incorrect.

I We can just be confident that we obtained a correct interval


(because 95% of the intervals we could have obtained are
correct).

I The greater the confidence level, the more confident we can


be that the confidence interval contains the actual population
parameter.
Confidence Interval for population mean
(A). CI for µ when σ 2 is known
I Theorem 1:

Assume that X1 , X2 , . . . , Xn is a random sample from a normal


population with mean µ and variance σ 2 (known), so that,

σ2
 
X −µ
X ∼ N µ, and Z= √ ∼ N(0, 1).
n σ/ n

Then, a (1 − α) 100% CI for the mean µ is:


  σ   σ 
x − zα/2 √ , x + zα/2 √ .
n n

This interval is often referred to as the Z -interval for a mean.


(A). CI for µ when σ 2 is known

I Proof:

From the standard normal curve, we can see that:


 
P −zα/2 ≤ Z ≤ zα/2 = 1 − α.

Then, simply replacing Z , we get:


 
X−µ
P −zα/2 ≤ √ ≤ zα/2 = 1 − α.
σ/ n
Now, consider the inequality inside the brackets:
X −µ
−zα/2 ≤ √
σ/ n
≤ zα/2
   
σ σ
−zα/2 √ ≤ X −µ ≤ +zα/2 √
n n
   
σ σ
−X − zα/2 √ ≤ −µ ≤ −X + zα/2 √
n n
   
σ σ
X − zα/2 √ ≤ µ ≤ X + zα/2 √
n n

Thus, we obtain that the following probability statement is


true:
    
σ σ
P X − zα/2 √ ≤ µ ≤ X + zα/2 √ = 1 − α,
n n

and we can be a (1 − α) 100% confident that the mean µ is


in the interval:
    
σ σ
Example 1

A manufacturing company produces light bulbs, and they


claim that their new production process results in
longer-lasting bulbs than the old process.

In a random sample of 200 bulbs from the new process, the


average lifespan is found to be 1500 hours with a standard
deviation of 100 hours.

From the past data, it is known that bulbs from the old
process had an average lifespan of 1400 hours.

Based on a 95% confidence level, Is there enough evidence to


support the company’s claim that the new production process
produces longer-lasting light bulbs?
Solution

Substituting the values (x̄ = 1500, n = 200, σ = 100 , and


z0.025 = 1.96) into the formula for a Z -interval for a mean, we get:
    
100 100
1500 − 1.96 √ , 1500 + 1.96 √ .
200 200
Simplifying, we get a 95%Cl for the average lifetime of the light
bulbs from the new process: [1486.14 – 1513.86], i.e., we can be
95% confident that the average lifetime of the light bulbs from the
new process is between 1486.14 hours and 1513.86.
Since the interval does not contain the value 1400, there is
convincing evidence that the new production process produces
longer-lasting light bulbs.
Example 2

We have collected 12 measurements of the pH level in a particular


lake over a month. The pH measurements are as follows:

Measurement 1 2 3 4 5 6 7 8 9 10 11 12
pH Level 7.2 7.0 7.1 7.3 7.4 7.2 7.1 7.0 7.3 7.2 7.1 7.4

Assuming that the pH levels follow a normal distribution with a


standard deviation (σ) of 0.1, find the 95% confidence interval for
the mean pH level (µ) in the lake.
Solution

Given: n = 12, σ = 0.1, and x̄ = 7.192. Also, for a 95%Cl,

zα/2 = z0.025 = 1.96.

Now, the resulting 95%Cl is:


   
σ σ
x̄ − zα/2 √ ≤ µ ≤ x̄ + zα/2 √
n n
0.1 0.1
7.192 − 1.96 √ ≤ µ ≤ 7.192 + 1.96 √
12 12
7.135 ≤ µ ≤ 7.249

Thus, based on the sample data, the mean pH value in the lake is
[7.135, 7.249].
Length of an interval

If a CI for a parameter θ is

L<θ<U

then the length of the interval is

U − L.

I We are interested to obtain confidence intervals that are as


narrow as possible. For instance, which is a better option?
(a) We can be 90% confident that the average selling price of the
smartphone is between 20, 000 and 28, 000 INR.
(b) We can be 90% confident that the average selling price of the
smartphone is between 23, 000 and 25, 000 INR.
Length of an interval contd.

I In case of Z -interval,
     
σ σ
Length = x + zα/2 √ − x − zα/2 √
n n
 
σ
= 2zα/2 √ .
n

I Hence, the length of the Z -interval for a mean depends upon


3 factors:
the sample size n, the population standard deviation σ, and
the confidence level

I 95% is the most commonly used confidence level.


Determining Sample Size

Based on the formula for the Z -interval:


 
σ
x ± zα/2 √ ,
n

we say that,

(i) x is a point estimate of µ;


 
(ii) x ± zα/2 √σn is an interval estimate of µ;

(iii) √σ
is the standard error of the mean.
n
 
(iv) zα/2 √σn is the margin of error (denoted by E)

I Thus, if W is the width of the CI then E = W /2.


Thank you for listening!

You might also like