SA1 Tutorial I

Manjusha Kancharla

22/09/2019
Agenda

Review and Practice on:

I Normal Distribution Based Probabilities

ET
I Interval Estimation: Confidence Intervals for Population Mean
(Unknown population variance)
I Interval Estimation: Confidence Intervals for Population
Proportion
I Central Limit Theorem (CLT)
Practice Problem-1: Warm Up!!

The time required for putting together a food order at a restaurant

is normally distributed with µ = 45 min and ‡ = 8 min. The
restaurant manager plans to have work begin on an order 10 minutes
after the order has been placed and the customer is told that their
food will be ready within 1 hour from order. What is the probability
that the restaurant manager cannot meet his commitment?
Xnn
Practice Problem-1: Solution

Random
T
Variable
I
I 0
Let T be the time it takes to work on an order
We have that T ≥ N(45, 82 )
I
I
The work begins 10 minutes after getting the order
So we need the order to be completed in t Æ 50 minutes
tea patient
0 q value
T
Fun Question: Why am I using ‘t’ instead of ‘T’ here?
0 O of T
Practice Problem-1: Solution

What is the probability that the restaurant manager cannot meet his
commitment?

I Convert this into math!

Practice Problem-1: Solution
Normdistt
What is the probability that the restaurant manager cannot meet his

4
commitment?

350845
TfL
I Convert this into math!

D
I P(T Ø 50)
P
What do we do next?
700 625
P 2
 Practice Problem-1: Solution
mean
P Bin not np
variance
nPu P
Ti np
Nco D
ftp.p
p
 Practice Problem-1: Solution

own
conf Int for J Triuman

nr
I u En i En
s
µ X _th i
Practice Problem-1: Solution

What is the probability that the restaurant manager cannot meet his
commitment?

I Convert this into math!

I P(T Ø 50)

What do we do next?

I Convert to Z (Standard Normal)

I
I
Z = T ‡≠µ ≥ N(0, 1)
P(T Ø 50) = P(Z Ø 50≠45
e
8 ) = P(Z Ø 0.625)

3
I In R try function = pnorm(q=0.625, lower.tail = F)
I Or =1 – pnorm(q=0.625)
Practice Problem-1: Solution

What is the probability that the restaurant manager cannot meet his
commitment?

a
P(T Ø 50) = 0.2659
Practice Problem-2

7
A book publisher monitors the size of shipments of its textbooks to
university bookstores. For a sample of texts used at various schools,
the 95% confidence interval for the size of the shipment was 250 ±
45 books. Which, if any, of the following interpretations of this
interval, are correct?

1. All shipments are between 205 and 295 books

2. 95% of shipments are between 205 and 295 books.
3. The procedure that produced this interval generates ranges
that hold the population mean for 95% of samples.
4. If we get another sample, then we can be 95% sure that the
mean of this second sample is between 205 and 295.
5. We can be 95% confident that the range of 160 to 340 holds
the population mean.
Practice Problem-2: Solution

All shipments are between 205 and 295 books

I Incorrect. The interval describes, with 95% confidence, the

location of the average shipment size µ, not the sizes of
individual shipments

95% of shipments are between 205 and 295 books

I Incorrect. The interval does not describe individual shipments

The procedure that produced this interval generates ranges that

hold the population mean for 95% of samples.

I Correct. 95% of intervals created in this fashion contain the

true population mean.
Practice Problem-2: Solution

If we get another sample, then we can be 95% sure that the mean
of this second sample is between 205 and 295.

I Incorrect. The interval does not describe the mean of another

sample.

We can be 95% confident that the range of 160 to 340 holds the
population mean.

I Incorrect. The interval does not correspond to a 95%

confidence level.
Practice Problem-3

A survey of 5,250 business travelers worldwide conducted by OAG

Business Travel Lifestyle indicated that 91% of business travelers

o
consider legroom the most important in-flight feature. (Angle of
seat recline and food service were second and third, respectively.)
Give a 95% confidence interval for the proportion of all business
travelers who consider legroom the most important feature.
Practice Problem-3: Solution

Identify the random variable here.

Practice Problem-3: Solution

tpah
Pop

Sample
Identify the random variable here.
Proportion of business travelers who consider leg-room to be the
most important in-flight feature.
poop
t
p
Practice Problem-3: Solution

We want a 95% CI for a proportion.

I Let fi be the population proportion

I
I An estimate is the sample proportion, p =?
oval
 Practice Problem-3: Solution

540
2 We want a 95% CI for a proportion.

I Let fi be the population proportion

I An estimate is the sample proportion, p = 0.91

The (1 ≠ –) *100% CI for fi is :


p(1 ≠ p)
p ± z–/2 Ô
n
Practice Problem-3: Solution

The (1 ≠ –) *100% CI for fi is :


p(1 ≠ p)
p ± z–/2 Ô
n

I – = 0.05, z0.025 = 1.96, n = 5250

as

0.91 ú (1 ≠ 0.91)
0.91 ± 1.96 ú Ô = (0.90, 0.92)
5250
 Practice Problem-4 (a)

r
The owner of a Pizza delivery restaurant is concerned about delivery
times. He kept close track of the delivery times for a random
sample of 196 pizza deliveries and found that the sample mean and
standard deviation are 21 minutes and 4 minutes respectively. He

I
also found that 28 of these pizzas took over 30 minutes to deliver,
which were to be given away for free.
Question: What should be the owner’s estimate of the average
time taken to deliver pizza? (assume 95% confidence level)

1. [20.44, 21.56]

07
2. [20.34, 21.66]
3. [20.04, 21.96]
4. [19.44, 22.56]
 Practice Problem-4 (a): Solution

n 196
I 21 5 4
time to deliver
a
pizza
er true avg
G MY.CI IE

109722
195 0.025
 Practice Problem-4 (a): Solution

We have:

I i
t
I x̄ = 21 minutes.
I s = 4 minutes (sample standard deviation).
I n = 196

coil I Define µ as the true average time to deliver a pizza

Then, the (1 ≠ –) *100% CI for µ is :

s
x̄ ± tn≠1,–/2 ◊ Ô
n

we xTE
Practice Problem-4 (a): Solution

We have:

I x̄ = 21 minutes.
I s = 4 minutes (sample standard deviation).
I n = 196
I x̄ ± tn≠1,–/2 ◊ Ôsn
I – = 0.05
I tn≠1,–/2 = t196≠1,0.025 ??

In R:

[1] 1.972204
Practice Problem-4 (a): Solution

We have our confidence interval as:

Fa
If
4
21 ± 1.9722 ◊ Ô
196
[20.43651, 21.56349] ¥ [20.44, 21.56]

0
## Practice Problem-4 (b)
What fraction of pizzas are being given away for free? (assume 90%
confidence level)

0
1. [0.1, 0.18]
2. [0.05, 0.23]
3. [0.12, 0.16]
28 of 196 pizzas
4. [0.01, 0.18]
were delivered after
30 minutes
Practice Problem-4 (b): Solution

Identify the random variable here.

Proportion pizzas that are being

of
given away for free
IT
Practice Problem-4 (b): Solution

Identify the random variable here.

Proportion of pizzas that are being given away for free.
 Practice Problem-4 (b): Solution

We want a 95% CI for a proportion.

f 28
96
I Let fi be the population proportion
I An estimate is the sample proportion, p =?
1
001429
T mean t 0 th i
STA
Practice Problem-4 (b): Solution 10
A
The (1 ≠ –) *100% CI for fi is :

p(1 ≠ p)
p ± z–/2 Ô
n

I – = 0.1, z0.05 = 1.6448, n = 196

O

0.1429 ú (1 ≠ 0.1429)
0.1429 ± 1.6448 ú Ô = (0.1017, 0.1840)
196
Practice Problem-5
withdrawal amount
in a transaction
N L 00
Auditors at a small community bank randomly sample 100
withdrawal transactions made during the week at an ATM machine 50
located near the bank’s main branch. Over the past 2 years, the
average withdrawal amount has been $50 with a standard deviation S 40
of $40. Since audit investigations are typically expensive, the
auditors decide to not initiate further investigations if the mean
transaction amount of the sample is between $45 and $55. What is
the probability that in any given week, there will be an investigation?

55
P 45 E I E
I
Practice Problem-5: Solution

We have:

I X: Withdrawal amount
I µ = 50, ‡ = 40, n = 100

We are interested in the probability of an investigation. Which is

same as the probability that the sample mean is beyond $45 and
$55.
How would you go about finding this? What do we need to find a
probability?
Practice Problem-5: Solution

We have:

I X: Withdrawal amount
I µ = 50, ‡ = 40, n = 100

We are interested in the probability of an investigation. Which is

same as the probability that the sample mean is beyond $45 and
$55.
How would you go about findin this? What do we need to find a
probability?

I A probability distribution!
I We don’t have one!
Practice Problem-5: Solution

CLT to the rescue.

The distribution of the sample mean:

I will be normal when the distribution of data in the population

is normal
I will be approximately normal even if the distribution of data in
the population is not normal, under some conditions:

caraghst
I Each data point in the sample is independent of the other
I The sample size is large enough (30 usually considered large
enough) e
I If data is quite symmetric and has few outliers, even smaller
samples are fine. Otherwise, we need larger samples
Practice Problem-5: Solution
µ Pop
If these conditions hold:
YY
‡2 a
a
of
0X̄ ≥ N(µ,

I The standard deviation here, Ô‡ ,

n
)

is called the Standard Error

n
of the Mean, SE (X̄ ).
Practice Problem-5: Solution

it
4
d
2
I From Central Limit Theorem, we know that X̄ ≥ N(µ, ‡n ).
I The SE (X̄ ) = Ô‡ = 4
I X̄ ≥ N(50, 42 )Oon

7452
I Find P(45 Æ X̄ Æ 55) = P(≠1.25 Æ Z Æ 1.25) = 0.79
I But we want probability outside of this range 0
I Probability that the sample mean is beyond $45 and $55 is 1-
0.79 = 0.21!

45 125
5 41
Pop ravana known CIjamian
use z value

unknown no matter
Popp Variance t
use
the sample size

CI for IT pop proponon

always 2 value dist