Vous êtes sur la page 1sur 88

Hypothesis

Testing
Variance known?

Sampling Distribution

Over-the-counter stock selling prices


calculate average price of all stocks listed []

take a sample of 25 stocks and record price


calculate average price of the 25 stocks [x-bar]

take all possible samples of size 25


would all x-bars be equal?

average all the possible x-bars equals

Sampling Distribution

20

= 50

H0
Levine, Prentice-Hall

Sample Mean

Sampling Distribution
It is unlikely
that we would
get a sample
mean of this
value ...

20

= 50

H0
Levine, Prentice-Hall

Sample Mean

Sampling Distribution
It is unlikely
that we would
get a sample
mean of this
value ...
... if in fact this were
the population mean
20

= 50

H0
Levine, Prentice-Hall

Sample Mean

Sampling Distribution
It is unlikely
that we would
get a sample
mean of this
value ...

... therefore,
we reject the
hypothesis
that = 50.
... if in fact this were
the population mean

20

= 50

H0
Levine, Prentice-Hall

Sample Mean

Null Hypothesis

What is tested

Always has equality sign: , or

Designated H0
Example ... H0: 3

Levine, Prentice-Hall

Alternative
Hypothesis

Opposite of null hypothesis

Always has inequality sign: ,, or

Designated H1

Example
H1: < 3

Levine, Prentice-Hall

Decision

Reject null hypothesis

Retain, or, fail to reject, null


hypothesis

Do not use the term accept

Levine, Prentice-Hall

p-value

Probability of obtaining a test statistic more


extreme (or than actual sample value
given H0 is true

Called observed level of significance


Smallest value of H0 can be rejected

Used to make rejection decision


If p-value , reject H0
Levine, Prentice-Hall

Level of
Significance

Defines unlikely values of sample


statistic if null hypothesis is true
Called rejection region of sampling
distribution

Designated (alpha)
Typical values are .01, .05, .10

Selected by researcher at start


Levine, Prentice-Hall

Rejection Region (one-tail


test)
Sampling Distribution

Level of Confidence

Rejection
Region
1-

Nonrejection
Region

Critical
Value

Ho
Value
Levine, Prentice-Hall

Sample Statistic

Rejection Region (one-tail


test)
Sampling Distribution

Level of Confidence

Rejection
Region
1-

Nonrejection
Region

Critical
Value

Ho
Value

Sample Statistic

Observed
sample statistic
Levine,
Prentice-Hall

Rejection Region (one-tail


test)
Sampling Distribution

Level of Confidence

Rejection
Region

Critical
Observed
sample statistic Value

1-
Nonrejection
Region

Ho
Value
Levine, Prentice-Hall

Sample Statistic

Rejection Regions (twotailed test)


Sampling Distribution

Level of Confidence

Rejection
Region
1/2

Rejection
Region
1-

Nonrejection
Region

Critical
Value

Ho
Value

Critical
Value
Levine, Prentice-Hall

1/2

Sample Statistic

Rejection Regions (twotailed test)

Sampling Distribution

Level of Confidence

Rejection
Region
1/2

Rejection
Region
1-

Nonrejection
Region

Critical
Value

Ho
Value

Critical
Value
Levine, Prentice-Hall

1/2

Sample Statistic
Observed
sample statistic

Rejection Regions (twotailed test)

Sampling Distribution

Level of Confidence

Rejection
Region
1/2

Rejection
Region
1-

Nonrejection
Region

Critical
Value

Ho
Value

Critical
Value
Observed
Levine, Prentice-Hall
sample statistic

1/2

Sample Statistic

Rejection Regions (twotailed test)

Sampling Distribution

Level of Confidence

Rejection
Region
1/2

Observed
sample statistic

Rejection
Region
1-

Nonrejection
Region

Critical
Value

Ho
Value

Critical
Value
Levine, Prentice-Hall

1/2

Sample Statistic

Risk of Errors in Making


Decision

Type I error

Reject true null hypothesis


Has serious consequences
Probability of Type I error is alpha [
Called level of significance

Type II error

Do not reject false null hypothesis


Probability of Type II error is beta [
Levine, Prentice-Hall

Decision Results
H0: Innocent
Jury Trial

H0 Test

Actual Situation
Verdict
Innocent

Guilty

Innocent

Guilty

Correct

Error

Error

Correct

Actual Situation
Decision

H0 True

H0 False

Do Not
Reject H0

Correct

Type II
Error ( )

Type I
Error ()

Correct

Reject H0

Levine, Prentice-Hall

Hypothesis Testing

State H0

State H1

Choose

Choose n

Choose test

Levine, Prentice-Hall

Hypothesis Testing

State H0

Set up critical values

State H1

Collect data

Choose

Compute test statistic

Choose n

Make statistical decision

Choose test

Express decision

Levine, Prentice-Hall

Two-tailed z-test
Does an average box of
cereal contain 368 grams of
cereal? A random sample of
25 boxes has an average
weight = 372.5 grams. The
company has specified to
be 15 grams. Test at the .05
level.
368 gm.
Levine, Prentice-Hall

Two-tailed z-test
Test Statistic:
H0:
H1:

n
Critical Value(s):

Decision:
Conclusion:

Levine, Prentice-Hall

Two-tailed z-test
H0: = 368
H1: 368

n
Critical Value(s):

Test Statistic:

Decision:
Conclusion:

Levine, Prentice-Hall

Two-tailed z-test
H0: = 368
H1: 368
.05
n 25
Critical Value(s):

Test Statistic:

Decision:
Conclusion:

Levine, Prentice-Hall

Two-tailed z-test
H0: = 368
H1: 368
.05
n 25
Critical Value(s):
Reject H

Reject H

.025
-1.96

.025
0 1.96

Test Statistic:

Decision:
Conclusion:

Levine,
Prentice-Hall
Z

Two-tailed z-test
Test Statistic:

H0: = 368
X 372.5 368
1.50
H1: 368 Z
15
.05
n
25
n 25
Decision:
Critical Value(s):
Reject H

00

Reject H

.025
-1.96

.025
0 1.96

00

Conclusion:

Levine,
Prentice-Hall
Z

Two-tailed z-test
Test Statistic:

H0: = 368
X 372.5 368
1.50
H1: 368 Z
15
.05
n
25
n 25
Decision:
Critical Value(s):
Do not reject at = .05
Reject H

.025
-1.96

00

Reject H

.025

00

Conclusion:

Prentice-Hall
0 1.96 Levine,
Z

Two-tailed z-test
Test Statistic:

H0: = 368
X 372.5 368
1.50
H1: 368 Z
15
.05
n
25
n 25
Decision:
Critical Value(s):
Do not reject at = .05
Reject H

.025
-1.96

00

Reject H

.025

00

Conclusion:
No evidence
average is not 368

Prentice-Hall
0 1.96 Levine,
Z

Two-tailed z-test

-1.50

1.50

[p-value]]

Z value of sample
statistic
Levine, Prentice-Hall

Two-tailed z-test

[p-value]

p-value is P(z -1.50 or z 1.50)

-1.50

1.50

Z value of sample
statistic
Levine, Prentice-Hall

Two-tailed z-test

[p-value]

p-value is P(z -1.50 or z 1.50)

1/2 p-Value

1/2 p-Value

-1.50

1.50

Z value of sample
statistic
Levine, Prentice-Hall

Two-tailed z-test

[p-value]

p-value is P(z -1.50 or z 1.50)

1/2 p-Value

1/2 p-Value
.4332

-1.50

1.50

From Z table:
Z value of sample
lookup 1.50
statistic
Levine, Prentice-Hall

Two-tailed z-test

[p-value]

p-value is P(z -1.50 or z 1.50)

1/2 p-Value
.0668

1/2 p-Value
.0668

.5000
- .4332
.0668

.4332

-1.50

1.50

From Z table:
Z value of sample
lookup 1.50
statistic
Levine, Prentice-Hall

Two-tailed z-test

[p-value]

p-value is P(z -1.50 or z 1.50) = .1336

1/2 p-Value
.0668

1/2 p-Value
.0668

.5000
- .4332
.0668

.4332

-1.50

1.50

From Z table:
Z value of sample
lookup 1.50
statistic
Levine, Prentice-Hall

Two-tailed z-test
1/2 p-value = .0668

[p-value]

1/2 p-value = .0668

Reject

Reject

1/2 = .025

1/2 = .025

-1.50

1.50

Levine, Prentice-Hall

Two-tailed z-test

[p-value]

(p-Value = .1336) ( = .05)


1/2 p-Value = .0668

Do not reject.

1/2 p-Value = .0668

Reject

Reject

1/2 = .025

1/2 = .025

-1.50

1.50

Test statistic is in Do not reject region


Levine, Prentice-Hall

Two-tailed z-test ( known)


challenge
You are a Q/C inspector. You want to find
out if a new machine is making electrical
cords to customer specification: average
breaking strength of 70 lb. with = 3.5 lb.
You take a sample of 36 cords & compute
a sample mean of 69.7 lb. At the .05
level, is there evidence that the machine
is not meeting the average breaking
strength?

Levine, Prentice-Hall

Two-tailed z-test
known)

Test Statistic:
H0:
H1:
=
n=
Critical Value(s):

Decision:
Conclusion:

Levine, Prentice-Hall

Two-tailed z-test
known)

Test Statistic:
H0: = 70
H1: 70
=
n=
Critical Value(s):

Decision:
Conclusion:

Levine, Prentice-Hall

Two-tailed z-test
known)

Test Statistic:
H0: = 70
H1: 70
= .05
n = 36
Critical Value(s):

Decision:
Conclusion:

Levine, Prentice-Hall

Two-tailed z-test
known)

Test Statistic:
H0: = 70
H1: 70
= .05
n = 36
Critical Value(s):
Reject H

.025
-1.96

00

Reject H

.025

00

Decision:
Conclusion:

0 1.96 Levine,
Z
Prentice-Hall

Two-tailed z-test
known)

Test Statistic:
H0: = 70
X 69.7 70
Z

.51
H1: 70

3.5
= .05
n
36
n = 36
Decision:
Critical Value(s):
Reject H

Reject H

.025
-1.96

.025
0 1.96

Conclusion:

Z
Levine,
Prentice-Hall

Two-tailed z-test
known)

Test Statistic:

Reject H

H0: = 70
X 69.7 70
Z

.51
H1: 70

3.5
= .05
n
36
n = 36
Decision:
Critical Value(s):
Do not reject at = .05
0

Reject H

.025
-1.96

.025
0 1.96

Conclusion:

Z
Levine,
Prentice-Hall

Two-tailed z-test
known)

Test Statistic:

Reject H

H0: = 70
X 69.7 70
Z

.51
H1: 70

3.5
= .05
n
36
n = 36
Decision:
Critical Value(s):
Do not reject at = .05
0

Reject H

.025
-1.96

.025
0 1.96

Conclusion:
No evidence
average is not 70

Z
Levine,
Prentice-Hall

One-tailed z-test
known)

Assumptions
Population is normally distributed
If not normal, can be approximated by
normal distribution for large samples

Levine, Prentice-Hall

One-tailed z-test
known)

Assumptions
Population is normally distributed
If not normal, can be approximated by
normal distribution for large samples

Null hypothesis has or sign only

Levine, Prentice-Hall

One-tailed z-test
known)

Assumptions
Population is normally distributed
If not normal, can be approximated by
normal distribution for large samples

Null hypothesis has or sign only


Z-test statistic

X x
X
Z

x
Levine, Prentice-Hall
n

One-tailed z-test
H0:0 H1: < 0

Reject H

Must be significantly
below
Levine, Prentice-Hall

( known)

One-tailed z-test
H0:0 H1: < 0

Reject H

( known)

H0:0 H1: > 0

Reject H

0
Must be significantly
below

0
Small values satisfy
H0 . Do not reject!

Levine, Prentice-Hall

One-tailed z-test
What is z given = .025?

=1
= .025

Levine, Prentice-Hall

( known)

One-tailed z-test
What Is Z given = .025?
.500
- .025
.475

=1
= .025

Levine, Prentice-Hall

( known)

One-tailed z-test
What is z given = .025?

.500
- .025
.475

=1

( known)

Standardized Normal
Probability Table (Portion)

.05

.06

.07

1.6 .4505 .4515 .4525

= .025

1.7 .4599 .4608 .4616


1.8 .4678 .4686 .4693

1.9

.4744 .4750

Levine, Prentice-Hall

.4756

One-tailed z-test
What Is Z given = .025?
.500
- .025
.475

=1

( known)

Standardized Normal
Probability Table (Portion)

.05

.06

.07

1.6 .4505 .4515 .4525

= .025

0 1.96

1.7 .4599 .4608 .4616


1.8 .4678 .4686 .4693

1.9

.4744 .4750

Levine, Prentice-Hall

.4756

One-tailed z-test
Does an average box of
cereal contain more than
368 grams of cereal? A
random sample of 25
boxes showed X = 372.5.
The company has
specified to be 15 grams.
Test at the .05 level.
Levine, Prentice-Hall

( known)

368 gm.

One-tailed z-test
known)

Test Statistic:
H0:
H1:
=
n=
Critical Value(s):

Decision:
Conclusion:

Levine, Prentice-Hall

One-tailed z-test
known)

Test Statistic:
H0: 368
H1: > 368
=
n=
Critical Value(s):

Decision:
Conclusion:

Levine, Prentice-Hall

One-tailed z-test
known)

Test Statistic:
H0: 368
H1: > 368
= .05
n = 25
Critical Value(s):

Decision:
Conclusion:

Levine, Prentice-Hall

One-tailed z-test
known)

Test Statistic:
H0: 368
H1: > 368
= .05
n = 25
Critical Value(s):

Decision:

Reject

.05
0 1.645

Conclusion:

Z
Levine,
Prentice-Hall

One-tailed z-test
known)

Test Statistic:
H0: 368
X 372.5 368
1.50
H1: > 368 Z
15
= .05
n
25
n = 25
Decision:
Critical Value(s):
Reject

.05
0 1.645

Conclusion:

Z
Levine,
Prentice-Hall

One-tailed z-test
known)

Test Statistic:
H0: 368
X 372.5 368
1.50
H1: > 368 Z
15
= .05
n
25
n = 25
Decision:
Critical Value(s):
Do not reject at = .05
Reject

.05
0 1.645

Conclusion:

Z
Levine,
Prentice-Hall

One-tailed z-test
known)

Test Statistic:
H0: 368
X 372.5 368
1.50
H1: > 368 Z
15
= .05
n
25
n = 25
Decision:
Critical Value(s):
Do not reject at = .05
Reject

.05
0 1.645

Conclusion:
No evidence average
is more than 368
Levine,
Prentice-Hall
Z

One-tailed z-test

known)

p-value Solution

Use
alternative
hypothesis
to find
direction

1.50

Z value of sample
statistic
Levine, Prentice-Hall

One-tailed z-test

p-value

known)

p-Value
.0668

Use
alternative
hypothesis
to find
direction

1.50

Z value of sample
statistic
Levine, Prentice-Hall

One-tailed z-test
known)

p-value
p-Value
.0668

Use
alternative
hypothesis
to find
direction

.4332

1.50

From Z table:
Z value of sample
lookup 1.50
statistic
Levine, Prentice-Hall

One-tailed z-test
p-value

known)

p-Value
.0668

Use
alternative
hypothesis
to find
direction

.5000
- .4332
.0668

.4332

1.50

From Z table:
Z value of sample
lookup 1.50
statistic
Levine, Prentice-Hall

One-tailed z-test

known)
p-value
p-value = .0668

p-Value
.0668

Use
alternative
hypothesis
to find
direction

.5000
- .4332
.0668

.4332

1.50

From Z table:
Z value of sample
lookup 1.50
statistic
Levine, Prentice-Hall

One-tailed z-test
known)

p-value
p-value = .0668

Reject
= .05

1.50

Levine, Prentice-Hall

One-tailed z-test
known)

p-value

(p-value = .0668) ( = .05).

Do not reject.p-Value = .0668

Reject
= .05

1.50

Test statistic is in Fail to reject region


Levine, Prentice-Hall

p-value Challenge
Youre an analyst for Ford. You want
to find out if the average miles per
gallon of Escorts is at least 32 mpg.
Similar models have a standard
deviation of 3.8 mpg. You take a
sample of 60 Escorts & compute a
sample mean of 30.7 mpg. What is
the value of the observed level of
significance (p-Value)?

Levine, Prentice-Hall

p-value
p-value = .004
p-value < ( = .01) Reject H0.

Use
alternative
hypothesis
to find
direction

p-Value
.004

.4960

-2.65

.5000
- .4960
.0040

Z value of
From Z table:
sample statistic
lookup 2.645
Levine, Prentice-Hall

p-value

Probability of obtaining a test statistic more


extreme (or than actual sample value
given H0 is true

Called observed level of significance


Smallest value of H0 can be rejected

Used to make rejection decision


If p-value , reject H0
Levine, Prentice-Hall

One-tailed t-test
unknown)

Does an average box of


cereal contain less than
the 368 grams indicated on
the package? A random
sample of 25 boxes
showed X = 363.5 and
s=15. Test at the .05 level.
368 gr.
Levine, Prentice-Hall

One-tailed t-test
unknown)

Test Statistic:
H0:
H1:
=
n=
Critical Value(s):

Decision:
Conclusion:

Levine, Prentice-Hall

One-tailed t-test
unknown)

Test Statistic:
H0: 368
H1: < 368
=
n=
Critical Value(s):

Decision:
Conclusion:

Levine, Prentice-Hall

One-tailed t-test
unknown)

Test Statistic:
H0: 368
H1: < 368
= .05
n = 25, d.f. = 24
Critical Value(s):

Decision:
Conclusion:

Levine, Prentice-Hall

One-tailed t-test
unknown)

Test Statistic:
H0: 368
H1: < 368
= .05
n = 25, d.f. = 24
Critical Value(s):

Decision:
Conclusion:

Levine, Prentice-Hall

One-tailed t-test
unknown)

Test Statistic:
H0: 368
X 363 . 5 368
t

1
.
50
H1: < 368
s
15
= .05
n
25
n = 25, d.f. = 24
Decision:
Critical Value(s):
Conclusion:
Levine, Prentice-Hall

One-tailed t-test
unknown)

Test Statistic:
H0: 368
X 363 . 5 368
t

1
.
50
H1: < 368
s
15
= .05
n
25
n = 25, d.f. = 24
Decision:
Critical Value(s):
Do not reject at = .05
Conclusion:
Levine, Prentice-Hall

One-tailed t-test
unknown)

Test Statistic:
H0: 368
X 363 . 5 368
t

1
.
50
H1: < 368
s
15
= .05
n
25
n = 25, d.f. = 24
Decision:
Critical Value(s):
Do not reject at = .05
Conclusion:
No evidence average
is less than 368
Levine, Prentice-Hall

One-tailed t-test

unknown)

p-value Solution

t value of sample
statistic
Levine, Prentice-Hall

Use
alternative
hypothesis
to find
direction

One-tailed t-test
unknown)

p-value

From t table: lookup


-1.50 for 24 d.f.

P-value
= 0.075

t value of sample
statistic
Levine, Prentice-Hall

Use
alternative
hypothesis
to find
direction

One-tailed t-test
unknown)

p-value

p-value = .075

= .05

Levine, Prentice-Hall

One-tailed t-test
unknown)

p-value

(p-value = .075) ( = .05).


p-value = .075

Do not reject.

Reject
= .05

Test statistic is in Fail to reject region


Levine, Prentice-Hall

Questions?

ANOVA

Vous aimerez peut-être aussi