stats_Chapter 8


View on GitHub | Download Local

Click to view slide text

Chapter 8 MATH 4753

By the end of this lesson you should know: •What hypothesis testing is. •What type 1 and 2 errors are. •What power is.

0.20 0.15 0.10 0.05 0.00

Development of Hypothesis testing (R script)

0

1

2

3

4

5

6

7

8

9

10

Test errors

•Type 1 error happens when you reject 𝑯𝒐 when it is true. –Probability of type 1 error =𝜶 •Type 2 error happens when you accept 𝑯𝒐 when it is false. –Probability of type 2 error=𝜷

The Courtroom

Power •Probability of rejecting 𝑯𝟎 when it is false is 1 − 𝛽 •This is called the power of a test. •Power is not the probability of an error! •Power=P(Rejecting 𝑯𝒐 |𝑯𝒐 is False)=1-P(Accepting 𝑯𝒐 |𝑯𝒐 is False)

What is the importance of power? •The higher the value of the power the more we can detect departures from 𝐻𝑜 .

Cut off and Acceptance, Rejection regions

Investigating POWER

10

12

10

16

18

H0

8

10

H1

12

14

x

x

xcut= 12

xcut= 14

H0

8

14

0.00 0.10 0.20 0.30

H1

H1

12 x

14

16

18

0.00 0.10 0.20 0.30

0.00 0.10 0.20 0.30

Density

8

Density

H0

xcut= 13

Density

0.00 0.10 0.20 0.30

Density

xcut= 11

𝐻0 : 𝜇 = 10 𝐻1 : 𝜇 = 15

H0

8

10

16

18

H1

12 x

14

16

18

Can you dig it?

•A) Yes •B) No

Environmental Protection Agency

Test

• 𝐻0 : 𝜇 = 3 • 𝐻1 : 𝜇 > 3

Power calculation

Hypothesis Testing: Errors

Use R

Population of Wisconsin lakes has 𝜇 = 15

qt(1-0.1/2, 25-1) [1] 1.710882

Probability the test will detect a mean that differs from 15 gm/m^3 if 𝜇𝑎 = 14

Taken from Montgomery and Runger App. Stat pg 311 Fourth edition

See BBD for Bayesian Example

𝑃 𝑥 =

𝑥

𝛼−1

1−𝑥 𝐵 𝛼, 𝛽

𝛽−1