0% found this document useful (0 votes)
34 views53 pages

5.confidence Interval

1) A confidence interval provides a range of values that is likely to include an unknown population parameter, with a specified degree of confidence. 2) The point estimate is the sample statistic used to estimate the population parameter. The confidence interval adds information about the variability or uncertainty of the estimate. 3) The formula for a confidence interval is: Point Estimate ± (Critical Value)(Standard Error). This gives the lower and upper confidence limits within which the population parameter lies with the stated confidence level.

Uploaded by

tunetonevidevide
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
0% found this document useful (0 votes)
34 views53 pages

5.confidence Interval

1) A confidence interval provides a range of values that is likely to include an unknown population parameter, with a specified degree of confidence. 2) The point estimate is the sample statistic used to estimate the population parameter. The confidence interval adds information about the variability or uncertainty of the estimate. 3) The formula for a confidence interval is: Point Estimate ± (Critical Value)(Standard Error). This gives the lower and upper confidence limits within which the population parameter lies with the stated confidence level.

Uploaded by

tunetonevidevide
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/ 53

21.11.

2020

Confidence Interval

§ A point estimate is a single number,


§ a confidence interval provides additional information about the
variability of the estimate

Lower Upper
Confidence Confidence
Point Estimate
Limit Limit
Width of
confidence interval

1
21.11.2020

We can estimate a with a Sample Statistic


Population Parameter (a Point Estimate)

Mean 𝜇 𝑋"

Standard Deviation 𝜎 𝑆

Proportion 𝜋 𝑝

§ How much uncertainty is associated with a point estimate of a


population parameter?

§ An interval estimate provides more information about a population


characteristic than does a point estimate

§ Such interval estimates are called “confidence intervals”

2
21.11.2020

§ An interval gives a range of values:


§ Takes into consideration variation in sample statistics from sample
to sample
§ Based on observations from one sample
§ Gives information about closeness to unknown population
parameters
§ Stated in terms of level of confidence
§ e.g. 95% confident, 99% confident
§ Can never be 100% confident

Cereal fill example


• Population has 𝜇 = 368 and 𝜎 = 15
• If you take a sample of size 𝑛 = 25 you know
!"
• 368 ∓ 1.96 ∗ = (362.12, 373.88) contains 95% of the sample means
#"
3 to estimate the population mean, 𝜇
• When you don’t know 𝜇, you use sample statistic “𝑋”
!"
• If 𝑋3 = 362.3, the interval for 𝜇 will be 362.3 ∓ 1.96 ∗ = (356.42, 368.18)
#"

• Since the actual mean is within this interval, this sample makes a correct statement about 𝜇

But what about the intervals from other possible samples of size 25?

3
21.11.2020

Lower Upper Contain


Sample # 𝑋"
Limit Limit µ?

1 362.30 356.42 368.18 Yes

2 369.50 363.62 375.38 Yes

3 360.00 354.12 365.88 No

4 362.12 356.24 368.00 Yes

5 373.88 368.00 379.76 Yes

§ In practice you only take one sample of size n


§ In practice you do not know µ so you do not know if the interval
actually contains µ
§ Howeveryou do know that 95% of the intervals formed in this
manner will contain µ
§ Thus, basedon the one sample, you actually selected you can be 95%
confident your interval will contain µ (this is a 95% confidence
interval)

4
21.11.2020

Random Sample
I am 95% confident
that μ is between 40
- 60.
Population Mean
(mean, μ, is X = 50
unknown)

Sample

§ The general formula for all confidence intervals is:

Point Estimate ± (Critical Value)(Standard Error)

10

5
21.11.2020

§ Confidence Level

§ Confidence the interval will contain the unknown population


parameter
§ A percentage (less than 100%)

11

§ Suppose confidence level = 95%

§ Also written (1 − a) = 0.95, (so a = 0.05)

§ A relative frequency interpretation:

§ 95% of all the confidence intervals


that can be constructed will
contain the unknown true parameter

§ A specific interval either will contain or will not contain the true
parameter

12

6
21.11.2020

Confidence Intervals
Population
Population Mean Proportion

𝜎 known 𝜎 unknown

13

§ Assumptions
§ Population standard deviation 𝜎 is known
§ Population is normally distributed
§ If population is not normal, use large sample

§ Confidence interval estimate:

𝜎
𝑋" ± 𝑍!/#
𝑛
where
𝑋3 is the point estimate
𝑍$# is the normal distribution critical value for a probability of a/2 in each tail
%
is the standard error
&

14

7
21.11.2020

§ Consider a 95% confidence interval: Zα/2 = ±1.96


1 - α = 0.95 so α = 0.05

α α
= 0.025 = 0.025
2 2

Z units: Zα/2 = -1.96 0 Zα/2 = 1.96


Lower Upper
X units: Confidence Point Estimate Confidence
Limit Limit

15

16

8
21.11.2020

§ Commonly used confidence levels are 90%, 95%, and 99%

Confidence
Confidence
Coefficient Zα/2 value
Level

80% 0.80 1.28


90% 0.90 1.645
95% 0.95 1.96
98% 0.98 2.33
99% 0.99 2.58
99.8% 0.998 3.08
99.9% 0.999 3.27

17

Sampling Distribution of the Mean

a/2 1- a a/2

Intervals
x
μx = μ
extend from x1
σ (1 − a)100%
x2
X - Zα / 2
of intervals constructed
to n
contain μ;
(a)100% do not.
σ
X + Zα / 2
n
Confidence Intervals

18

9
21.11.2020

§ A sample of 11 circuits from a large normal population has a mean


resistance of 2.20 ohms. We know from past testing that the population
standard deviation is 0.35 ohms.

§ Determine a 90% confidence interval for the true mean resistance of the
population.
𝜎
𝑋3 ± 𝑍$/#
𝑛

0.35
2.20 ± 1.645
11

2.026 ≤ 𝜇 ≤ 2.374

19

20

10
21.11.2020

§ We are 90% confident that the true mean resistance is between


2.026 and 2.374 ohms

§ Although the true mean may or may not be in this interval, 90%
of intervals formed in this manner will contain the true mean

21

§ A sample of 11 circuits from a large normal population has a mean


resistance of 2.20 ohms. We know from past testing that the population
standard deviation is 0.35 ohms.

§ Determine a 95% confidence interval for the true mean resistance of the
population.
𝜎
𝑋3 ± 𝑍$/#
𝑛

(.*"
2.20 ± 1.96 =2.20 ± 0.2068
!!

1.9932 ≤ 𝜇 ≤ 2.4068

22

11
21.11.2020

§ A sample of 11 circuits from a large normal population has a mean


resistance of 2.20 ohms. We know from past testing that the population
standard deviation is 0.35 ohms.

§ Determine a 99% confidence interval for the true mean resistance of the
population.
𝜎
𝑋3 ± 𝑍$/#
𝑛

0.35
2.20 ± 2.58
11

1.928 ≤ 𝜇 ≤ 2.472

23

24

12
21.11.2020

90% 95% 99%

2.026 ≤ 𝜇 ≤ 2.374 1.9932 ≤ 𝜇 ≤ 2.4068 1.928 ≤ 𝜇 ≤ 2.472

25

Confidence Intervals
Population
Population Mean Proportion

𝜎 known 𝜎 unknown

26

13
21.11.2020

§ Probably not!

§ In real world business situations, 𝜎 is not known.

a situation where 𝜎 is known then µ is also known (since to


§ If there is
calculate 𝜎 you need to know µ.)

§ If you truly know µ there would be no need to gather a sample to


estimate it.

27

§ If the population standard deviation 𝜎 is unknown, we can


substitute the sample standard deviation, S
§ This introduces extra uncertainty

§ So we use the t distribution instead of the normal distribution

28

14
21.11.2020

§ Assumptions
§ Population standard deviation 𝜎 is unknown
§ Population is normally distributed
§ If population is not normal, use large sample

§ Use Student’s t Distribution

§ Confidence interval estimate:

𝑆
𝑋" ± 𝑡!#
𝑛

where
𝑋 is the point estimate
𝑡$# is the critical value of t distribution with n-1 degrees of freedom and an area of α/2 in each tail
𝑆 is the standard deviation of the sample

29

§ The t is a family of distributions


§ The tα/2 value depends on degrees of freedom (d.f.)
§ Number of observations that are free to vary after
sample mean has been calculated
d.f. = n - 1

30

15
21.11.2020

Idea: Number of observations that are free to vary after sample mean has been
calculated
Example: Suppose the mean of 3 numbers is 8.0

Let X1 = 7 If the mean of these three values is


Let X2 = 8 8.0,
then X3 must be 9
What is X3?
(i.e., X3 is not free to vary)

Here, n = 3, so degrees of freedom = n – 1 = 3 – 1 = 2


Two values can be any numbers, but the third is not free to vary for a given mean

31

Note: t Z as n increases

Standard
Normal
(t with df = ∞)

t (df = 13)
t-distributions are bell-
shaped and symmetric,
but have ‘fatter’ tails than t (df = 5)
the normal

0 t

32

16
21.11.2020

Upper Tail Area


Let: n = 3
df = n - 1 = 2
df .10 .05 .025 a = 0.10
a/2 = 0.05
1 3.078 6.314 12.706

2 1.886 2.920 4.303

3 1.638 2.353 3.182 a/2 = 0.05

The body of the table


contains t values, not the
probabilities 0 2.920

33

Confidence t t t Z
Level (10 d.f.) (20 d.f.) (30 d.f.) (∞ d.f.)

0.80 1.372 1.325 1.310 1.28


0.90 1.812 1.725 1.697 1.645
0.95 2.228 2.086 2.042 1.96
0.99 3.169 2.845 2.750 2.58

34

17
21.11.2020

35

A random sample of 𝑛 = 25 has 𝑋+ = 50 and 𝑆 = 8.


Construct a 95% confidence interval for 𝜇.

d.f. = 𝑛 – 1 = 24, so 𝑡!/# = 𝑡$.$#& = 2.064

𝑆
𝑋3 ± 𝑡$/#
𝑛

8
50 ± 2.064
25

46.698 ≤ 𝜇 ≤ 53.302

36

18
21.11.2020

37

5, 7, 6, 8, 9, 11, 10, 9, 8, 7
Construct a 99% confidence interval for 𝜇.

d.f. = n – 1 = 9, so 𝑡!/# = 𝑡$.$$& = 3.25

80 (5 − 8)#+(7 − 8)#+(6 − 8)#+ ⋯ + (7 − 8)# 30


𝑋= =8 𝑆= = = 1.83
10 (10 − 1) 9

𝑆 1.83
𝑋3 ± 𝑡$/# 8 ± 3.25
𝑛 10

6.12 ≤ 𝜇 ≤ 9.88

38

19
21.11.2020

39

5, 8.5, 12, 15, 7, 9, 7.5, 6.5, 10.5


Construct a 95% confidence interval for 𝜇 if the population
variance is 9.

81 𝜎 3
𝑋" = =9 𝑋" ± 𝑍!/# 9 ± 1.96
9 𝑛 9

7.04 ≤ 𝜇 ≤ 10.96

40

20
21.11.2020

5, 8.5, 12, 15, 7, 9, 7.5, 6.5, 10.5


Construct a 95% confidence interval for 𝜇 if the population
variance is unknown .

81 (5 − 9)#+(8.5 − 9)#+(12 − 9)#+ ⋯ + (10.5 − 9)#


𝑋" = =9 𝑆= = 3.082
9 (9 − 1)

𝑆 3.082
𝑋" ± 𝑡!/# 9 ± 2.306 6.63 ≤ 𝜇 ≤ 11.37
𝑛 9
7.04 ≤ 𝜇 ≤ 10.96

41

42

21
21.11.2020

Confidence Intervals
Population
Population Mean Proportion

𝜎 known 𝜎 unknown

43

§ An interval estimate for the population proportion


(𝜋) can be calculated by adding an allowance for
uncertainty to the sample proportion (𝑝)

44

22
21.11.2020

§ Recall that the distribution of the sample proportion is


approximately normal if the sample size is large, with standard
deviation
𝜋(1 − 𝜋)
𝜎= =
𝑛
§ We will estimate this using sample data:

𝑝(1 − 𝑝)
𝑛

45

§ Upper and lower confidence limits for the population proportion are
calculated with the formula

𝑝(1 − 𝑝)
𝑝 ± 𝑍!/#
𝑛

§ where
§ 𝑍𝛼/2 is the standard normal value for the level of confidence desired
§ 𝑝 is the sample proportion
§ 𝑛 is the sample size

46

23
21.11.2020

§ A random sample of 100 people shows that 25 are left-handed.


§ Form a 95% confidence interval for the true proportion of left-
handers
𝑝(1 − 𝑝)
𝑝 ± 𝑍$/#
𝑛 We are 95% confident
that the true percentage
of left-handers in the
population is between
25 0.25(1 − 0.25) 16.51% and 33.49%
± 1.96
100 100

0.1651 ≤ 𝜋 ≤ 0.3349

47

Determining Sample Size

For Mean For Proportion

48

24
21.11.2020

§ The required sample size can be found to reach a desired


margin of error (𝑒) with a specified level of confidence (1 − a)
§ The margin of error is also called sampling error
§ the amount of imprecision in the estimate of the population
parameter
§ the amount added to and subtracted from the point estimate
to form the confidence interval

49

𝜎 𝜎 𝑍!/##𝜎 #
𝑋! ± 𝑍!/# 𝑒 = 𝑍!/# 𝑛=
𝑛 𝑛 𝑒#

50

25
21.11.2020

If s = 45, what sample size is needed to estimate the mean


within ± 5 with 90% confidence?

8-/. . 9. (;.<=>). =>.


𝑛= = = 219.19 ⟹ round up n=220
:. >.

51

§ If unknown, 𝜎 can be estimated when using the


required sample size formula
§ Use a value for 𝜎 that is expected to be at least as
large as the true 𝜎

§ Select a pilot sample and estimate 𝜎 with the sample


standard deviation, 𝑆

52

26
21.11.2020

Determining Sample Size

For Mean For Proportion

53

𝜋(1 − 𝜋) 𝑍!/##𝜋(1 − 𝜋)
𝑝 ± 𝑍A/C 𝑛=
𝑛 𝑒#

54

27
21.11.2020

How large a sample would be necessary to estimate the


true proportion defective in a large population within ±3%,
with 95% confidence?
Assume a pilot sample yields 𝑝 = 0.12

𝑍$/# # 𝜋(1 − 𝜋) 1.96 # 0.12(1 − 0.12)


𝑛= = = 450.74 ⟹ round up n=451
𝑒# (0.03)#

55

§ A confidence interval estimate (reflecting sampling error) should always be


included when reporting a point estimate
§ The level of confidence should always be reported

§ The sample size should be reported

§ An interpretation of the confidence interval estimate should also be provided

56

28
21.11.2020

57

§ FPC is used when we sample –without replacement- from more


than 5% of a finite population

§ It is needed because under these circumstances, the Central


Limit Theorem does not hold and the standard error of the
estimate will be too high.

§ FPC captures the difference between sampling with replacement


and sampling without replacement.

58

29
21.11.2020

§ Suppose 𝑛 = 100 is drawn from a population size 1000 with


standard deviation of 40.
40 40
𝜎? = = =4
100 10
@ ;BB
§ Since
A
= ;BBB = 0.10 > 0.05 and the population is finite,

§ Using FPC for the standard error of the mean, we get

40 𝑁 − 𝑛 40 1000 − 100
𝜎? = = = 3.8
100 𝑁 − 1 10 1000 − 1

59

§ Suppose 𝑛 = 100 is drawn from a population size 1000

§ 𝑋" = 50 and 𝑆 = 10. Construct 90% CI for mean.


@ ;BB
§ Since A = ;BBB = 0.10 > 0.05 and the population is finite,

§ Using FPC,

10 𝑁−𝑛
50 ± 1.984 = 50 ± 1.88
100 𝑁 − 1
48.12 ≤ 𝜇 ≤ 51.88

60

30
21.11.2020

61

Confidence Intervals
DEPENDENT
INDEPENDENT SAMPLES SAMPLES

𝜎; and 𝜎# are known 𝜎; and 𝜎# are unknown

𝜎C = 𝜎D 𝜎C ≠ 𝜎D

62

31
21.11.2020

INDEPENDENT SAMPLES
PARAMETERS POPULATION 1 POPULATION 2
MEAN 𝜇! 𝜇#
VARIANCE 𝜎!# 𝜎##
STANDARD DEVIATION 𝜎! 𝜎#

SAMPLE 1 SAMPLE 2
SIZE 𝑛! 𝑛#
MEAN 𝑋3! 𝑋3#
VARIANCE 𝑆!# 𝑆##
STANDARD DEVIATION 𝑆! 𝑆#

63

INDEPENDENT SAMPLES

𝐸 𝑋"; − 𝑋"# = 𝐸 𝑋"; - 𝐸 𝑋"# = 𝜇; − 𝜇#

𝜎;# 𝜎##
Var 𝑋"; − 𝑋"# = 𝑉𝑎𝑟 𝑋"; + 𝑉𝑎𝑟 𝑋"# = +
𝑛; 𝑛#

64

32
21.11.2020

INDEPENDENT SAMPLES
§ 120 marketing and 90 finance students are randomly selected from two
independent populations with 0.42 and 0.64 standard deviations,
respectively. The mean GPA of marketing students is 3.08 and finance
students is 2.88.
§ Costruct 95% confidence interval for the differences of population
means.

𝑛! = 120 𝑛# = 90

𝑋3! = 3.08 𝑋3# = 2.88

𝜎! = 0.42 𝜎# = 0.64

65

INDEPENDENT SAMPLES
𝑛! = 120 𝑛# = 90

𝑋3! = 3.08 𝑋3# = 2.88

𝜎! = 0.42 𝜎# = 0.64

𝜎;# 𝜎## 0.42# 0.64#


(𝑋";−𝑋"#) ± 𝑍!/# + (3.08 − 2.88) ± 1.96 +
𝑛; 𝑛# 120 90

0.0479 ≤ 𝜇; − 𝜇# ≤ 0.3521

66

33
21.11.2020

Confidence Intervals
DEPENDENT
INDEPENDENT SAMPLES SAMPLES

𝜎; and 𝜎# are known 𝜎; and 𝜎# are unknown

𝜎C = 𝜎D 𝜎C ≠ 𝜎D

67

INDEPENDENT SAMPLES
PARAMETERS POPULATION 1 POPULATION 2
MEAN 𝜇! 𝜇#
VARIANCE 𝜎!# 𝜎##
STANDARD DEVIATION 𝜎! 𝜎#

SAMPLE 1 SAMPLE 2
SIZE 𝑛! 𝑛#
MEAN 𝑋3! 𝑋3#
VARIANCE 𝑆!# 𝑆##
STANDARD DEVIATION 𝑆! 𝑆#

68

34
21.11.2020

INDEPENDENT SAMPLES
§ Since we assume 𝜎;= 𝜎# = 𝜎, we need to estimate a common variance
using sample standard deviations
§ If 𝑛; = 𝑛#, then
𝑆;# + 𝑆##
𝑆CDDE# =
2

§ If 𝑛; ≠ 𝑛#, then

(𝑛; − 1)𝑆;#+(𝑛# − 1)𝑆##


𝑆CDDE# = , 𝑡!/#,(@/ G@. H#)
(𝑛; + 𝑛# − 2)

69

INDEPENDENT SAMPLES
§ 25 marketingand 25 finance students are randomly selected from two
independent populations with equal variances. The mean GPA of
marketing students is 3.08 and finance students is 2.88.
§ Costruct 95% confidence interval for the differences of population
means.

𝑛! = 25 𝑛# = 25

𝑋3! = 3.08 𝑋3# = 2.88

𝑆! # = 0.36 𝑆# # = 0.64

70

35
21.11.2020

INDEPENDENT SAMPLES
𝑛! = 25 𝑛# = 25

𝑋3! = 3.08 𝑋3# = 2.88

𝑆! # = 0.36 𝑆# # = 0.64

𝑆CDDE# 𝑆CDDE# 0.36 + 0.64


(𝑋";−𝑋"#) ± 𝑡!/# + 𝑆CDDE# = = 0.05
𝑛; 𝑛# 2

(25 − 1)𝑆;#+(25 − 1)𝑆## 24(𝑆;# + 𝑆##) (𝑆;# + 𝑆##)


𝑆CDDE# = = =
(25 + 25 − 2) 48 2

71

72

36
21.11.2020

INDEPENDENT SAMPLES
𝑛! = 25 𝑛# = 25

𝑋3! = 3.08 𝑋3# = 2.88

𝑆! # = 0.36 𝑆# # = 0.64

0.05 0.05
(3.08 − 2.88) ± 2.02 + −0.20 ≤ 𝜇; − 𝜇# ≤ 0.60
25 25

73

INDEPENDENT SAMPLES
§ 25 marketing and 40 finance students are randomly selected from two
independent populations with equal variances. The mean GPA of
marketing students is 3.08 and finance students is 2.88.
§ Costruct 95% confidence interval for the differences of population
means.

𝑛! = 25
𝑛# = 40
𝑋3! = 3.08
𝑋3# = 2.88
𝑆! # = 0.36
𝑆# # = 0.64

74

37
21.11.2020

INDEPENDENT SAMPLES
𝑛! = 25 𝑛# = 40

𝑋3! = 3.08 𝑋3# = 2.88

𝑆! # = 0.36 𝑆# # = 0.64

𝑆CDDE# 𝑆CDDE#
(𝑋";−𝑋"#) ± 𝑡!/# +
𝑛; 𝑛#

# (25 − 1)𝑆;#+(40 − 1)𝑆## 24 0.36 + 39(0.64)


𝑆CDDE = = = 0.50
(25 + 40 − 2) 63

75

76

38
21.11.2020

INDEPENDENT SAMPLES
𝑛! = 25 𝑛# = 40

𝑋3! = 3.08 𝑋3# = 2.88

𝑆! # = 0.36 𝑆# # = 0.64

0.50 0.50
(3.08 − 2.88) ± 2.00 + −0.16 ≤ 𝜇; − 𝜇# ≤ 0.56
25 40

77

Confidence Intervals
DEPENDENT
INDEPENDENT SAMPLES SAMPLES

𝜎; and 𝜎# are known 𝜎; and 𝜎# are unknown

𝜎C = 𝜎D 𝜎C ≠ 𝜎D

78

39
21.11.2020

INDEPENDENT SAMPLES
PARAMETERS POPULATION 1 POPULATION 2
MEAN 𝜇! 𝜇#
VARIANCE 𝜎!# 𝜎##
STANDARD DEVIATION 𝜎! 𝜎#

SAMPLE 1 SAMPLE 2
SIZE 𝑛! 𝑛#
MEAN 𝑋3! 𝑋3#
VARIANCE 𝑆!# 𝑆##
STANDARD DEVIATION 𝑆! 𝑆#

79

INDEPENDENT SAMPLES
§ Since we assume 𝜎; ≠ 𝜎#, we cannot estimate a common variance
§ We use the sample standard deviations and calculate a common
degrees of freedom
#
𝑆;# 𝑆## Round to the

𝑛; + 𝑛#
nearest
integer!
df = # #
𝑆;# 𝑆##
𝑛; 𝑛#
+
(𝑛; − 1) (𝑛# − 1)

80

40
21.11.2020

INDEPENDENT SAMPLES
§ 50 marketing and 25 finance students are randomly selected from two
independent populations. The mean GPA of marketing students is 3.08
and finance students is 2.88.
§ Costruct 95% confidence interval for the differences of population
means.

𝑛! = 50 𝑛# = 25

𝑋3! = 3.08 𝑋3# = 2.88

𝑆! # = 0.36 𝑆# # = 0.64

81

INDEPENDENT SAMPLES
𝑛! = 50 𝑛# = 25

𝑋3! = 3.08 𝑋3# = 2.88

𝑆! # = 0.36 𝑆# # = 0.64

0.36 0.64 #
𝑆;# 𝑆## df = 50 + 25 = 37.93 ⟹ 38
(𝑋";−𝑋"#) ± 𝑡!/# +
𝑛; 𝑛# 0.36 # 0.64 #
50 + 25
(50 − 1) (25 − 1)

82

41
21.11.2020

83

INDEPENDENT SAMPLES
𝑛! = 50 𝑛# = 25

𝑋3! = 3.08 𝑋3# = 2.88

𝑆! # = 0.36 𝑆# # = 0.64

0.36 0.64
(3.08 − 2.88) ± 2.02 + −0.17 ≤ 𝜇; − 𝜇# ≤ 0.57
50 25

84

42
21.11.2020

EXERCISES
§ Anexperiment was conducted in which two types of engines, A and B,
were compared. Gas mileage, in miles per gallon, was measured.

§ 50experiments were conducted using engine type A and 75 were done


for engine type B. The gasoline used and other conditions were held
constant.

§ Theaverage mileage for engine A was 36 mpg and the average for
machine B was 42 mpg.

§ Findan approximate 96% CI on 𝜇I − 𝜇J . Standard deviation of the


population are 6 and 8 for machines A and B, respectively.

(Walpole et al., 2011; p. 286)

85

YES NO
Independent
Samples?

YES Do you
know the
population NO
variance?

YES Assume
equal
Z distribution variances? NO
Use 𝜎

t distribution
Calculate t distribution
common Calculate df.
variance, 𝑆0112 #

86

43
21.11.2020

YES NO
Independent
Samples?

YES Do you
know the
population NO
variance?

YES Assume
equal
Z distribution variances? NO
Use 𝜎

t distribution
Calculate t distribution
common Calculate df.
variance, 𝑆0112 #

87

EXERCISES
𝑛3 = 50 𝑛4 = 75

𝑋33 = 36 𝑋34 = 42

𝜎3 = 6 𝜎4 = 8

𝜎I # 𝜎J # 8# 6#
(𝑋"I −𝑋"J ) ± 𝑍!/# + (42 − 36) ± 2.05 +
𝑛I 𝑛J 75 50

3.43 ≤ 𝜇I − 𝜇J ≤ 8.57

88

44
21.11.2020

EXERCISES
𝑛! = 120 𝑛# = 90

𝑋3! = 3.08 𝑋3# = 2.88

𝜎! = 0.42 𝜎# = 0.64

𝜎;# 𝜎## 0.42# 0.64#


(𝑋";−𝑋"#) ± 𝑍!/# + (3.08 − 2.88) ± 1.96 +
𝑛; 𝑛# 120 90

0.0479 ≤ 𝜇; − 𝜇# ≤ 0.3521

89

EXERCISES
§ Two independentsampling stations were chosen for a study, one located
downstream from the acid mine discharge point and the other located
upstream.
§ For 12 monthly samples collected at the downstream station, the species
diversity index had a mean value 3.11 and a standard deviation 0.771,
while 10 monthly samples collected at the upstream station had a mean
index value 2.04 and a standard deviation 0.448.
§ Find
a 90% confidence interval for the difference between the population
means for the two locations, assuming that the populations are
approximately normally distributed with equal variances.

(Walpole et al., 2011; p. 288)

90

45
21.11.2020

YES NO
Independent
Samples?

YES Do you
know the
population NO
variance?

YES Assume
equal
Z distribution variances? NO
Use 𝜎

t distribution
Calculate t distribution
common Calculate df.
variance, 𝑆0112 #

91

EXERCISES
𝑛! = 12 𝑛# = 10

𝑋3! = 3.11 𝑋3# = 2.04

𝑆! = 0.771 𝑆# = 0.448

𝑆CDDE# 𝑆CDDE#
(𝑋";−𝑋"#) ± 𝑡!/# +
𝑛; 𝑛#

# #
11 0.771 + 9 0.448
𝑆CDDE# = = 0.417
20

92

46
21.11.2020

93

EXERCISES
𝑛! = 12 𝑛# = 10

𝑋3! = 3.11 𝑋3# = 2.04

𝑆! = 0.771 𝑆# = 0.448

0.417 0.417
(3.11 − 2.04) ± 1.725 + 0.59 ≤ 𝜇; − 𝜇# ≤ 1.55
12 10

94

47
21.11.2020

EXERCISES
§ A study was conducted by the Department of Zoology at the Virginia Tech to
estimate the difference in the amounts of the chemical orthophosphorus
measured at two different stations on the James River.

§ Orthophosphorus was measured in milligrams per liter. The 15 samples from


station 1 had an average orthophosphorus content of 3.84 milligrams per liter
and a standard deviation of 3.07 milligrams per liter, while the 12 samples from
station 2 had an average content of 1.49 milligrams per liter and a standard
deviation of 0.80 milligram per liter.

§ Find a 95% confidence interval for the difference in the true average
orthophosphorus contents at these two stations, assuming that the observations
came from normal populations with different variances.

(Walpole et al., 2011; p. 290)

95

YES NO
Independent
Samples?

YES Do you
know the
population NO
variance?

YES Assume
equal
Z distribution variances? NO
Use 𝜎

t distribution
Calculate t distribution
common Calculate df.
variance, 𝑆0112 #

96

48
21.11.2020

EXERCISES
𝑛! = 15 𝑛# = 12

𝑋3! = 3.84 𝑋3# = 1.49

𝑆! = 3.07 𝑆# = 0.80

#
3.07# 0.80#
𝑆;# 𝑆## + 12
15
(𝑋";−𝑋"#) ± 𝑡!/# + df = # # = 16.3 ⟹ 16
𝑛; 𝑛# 3.07# 0.80#
15 12
+
(15 − 1) (12 − 1)

97

98

49
21.11.2020

EXERCISES
𝑛! = 15 𝑛# = 12

𝑋3! = 3.84 𝑋3# = 1.49

𝑆! = 3.07 𝑆# = 0.80

3.07# 0.80#
(3.84 − 1.49) ± 2.12 + 0.60 ≤ 𝜇; − 𝜇# ≤ 4.10
15 12

99

Confidence Intervals
DEPENDENT
INDEPENDENT SAMPLES SAMPLES

𝜎; and 𝜎# are known 𝜎; and 𝜎# are unknown

𝜎C = 𝜎D 𝜎C ≠ 𝜎D

100

50
21.11.2020

DEPENDENT SAMPLES
§ We are often concerned with data sets that consist of pairs of values that
have some relationship to each other.

§ Paired
t-test, a special case of the two-sample t-tests, occurs when the
observations on the two populations of interest are collected in pairs.

101

YES NO
Independent
Samples? Paired t-test

YES Do you
know the
population NO
variance?

YES Assume
equal
Z distribution variances? NO
Use 𝜎

t distribution
Calculate t distribution
common Calculate df.
variance, 𝑆0112 #

102

51
21.11.2020

DEPENDENT SAMPLES
Observation Drug A Drug B Find 99% CI for the difference
1 29 26 of the population means

2 32 27
3 31 28
4 32 27
5 30 -
6 32 30
7 29 26
8 31 33
9 30 36

103

DEPENDENT SAMPLES
Observation Drug A Drug B
1 29 26
2 32 27
3 31 28
4 32 27
5 30 -
Missing value
6 32 30
7 29 26
8 31 33
9 30 36

104

52
21.11.2020

DEPENDENT SAMPLES
Observation Drug A Drug B Differences 13
1 29 26 𝑑= = 1.625
3 8
2 32 27 5
𝑆5 = 3.78
3 31 28 3
4 32 27 5
∑867!(𝑥6 − 𝑑)#
5 32 30 2 𝑆5 =
(𝑛 − 1)
6 29 26 3
7 31 33 -2
8 30 36 -6

105

DEPENDENT SAMPLES

𝑆P
𝜇̅P = 𝑑̅ ± 𝑡!/#
𝑛

Q.RS
𝜇̅P = 1.625 ± 3.499 S
−3.05 ≤ 𝜇̅P ≤ 6.30

106

53

You might also like