Math 201 - PowerPoint PPT Presentation

1 / 17
About This Presentation
Title:

Math 201

Description:

Confidence Interval for m. Significance test for m. Chapter 7 s is not known. 5. If s is Not Known? ... Z Test for Mean ( Known) Assumptions. Population is ... – PowerPoint PPT presentation

Number of Views:21
Avg rating:3.0/5.0
Slides: 18
Provided by: Pin112
Category:

less

Transcript and Presenter's Notes

Title: Math 201


1
Math 201
  • 6.2 Tests of Significance ctd
  • 7.1 Inferences on m

2
Two-sided Tests and Confidence Intervals
For x372.5, s10 and n25 ? sx10/52
95 CI for m is (372.5-1.962,
372.51.962)(368.58, 376.42)
H0 m 368 Ha m ¹ 368
Since 368 is outside the interval, we reject H0
at a .05.
3
Math 201
  • Chapter 7
  • 7.1 Inference for m

4
What is different?
  • Chapter 6 ? Assuming s known
  • Confidence Interval for m
  • Significance test for m
  • Chapter 7 ? s is not known

5
If s is Not Known?
  • Substitute the sample standard deviation for s

Standard error of X
6
Z Test for Mean( Known)
  • Assumptions
  • Population is normally distributed
  • If not normal, requires large samples
  • Z test statistic

Z distribution
7
t Test Unknown
  • Assumption
  • Population is normally distributed
  • If not normal, requires a large sample
  • T test

t distribution with n-1 degrees of freedom
8
t distribution
Table D
9
CI for ( not known)
  • Assumptions
  • Population is normally distributed
  • If population is not normal, use large sample
  • C Confidence interval for m

Estimate Margin of Error
Page 517 26
10
Estimation Process
Population All adult women
Random Sample n4
What the value of m?
64, 67, 61, 68 x 65
XHeight ? is unknown
Sample
11
95 Level of CI
t3 curve
.95
.025
.025
-t
t
x 65 s3.16 n4
(65-3.1821.58, 653.1821.58)
Exercise 7.10
( 59.97, 70.03)
Page 513 6
12
Guidelines
  • Practical guidelines for inference on m
  • nlt15 ? Use t if data are close to normal and no
    outliers.
  • n gt15 ? Use t if no outliers or strong skewness.
  • n gt40 ? Use t even for skewed distributions.

13
Matched Pair Design
H
T
H
PAIR
RANDOMIZE
TREAT
14
Matched pairs designs
  • Compares only two treatments
  • Choose two EUs that are as closely matched as
    possible
  • Assign treatments randomly to each EU.
  • Sometimes each pair consists of just one Subject.
  • Example Coke Vs Pepsi
  • ?Compare, randomization, repetition

15
Example
  • Weight loss program Here is an SRS of
    participants

Is the program effective?
16
Is the Program Effective?
  • Let m Ave wt. Loss for all participants
  • H0 m0
  • Ha mgt0
  • Minitab
  • Data for the sample of 8 participants
  • ? Normal?
  • ? average 8.94 lbs
  • standard deviation 10.47 lbs

CI?
17
HW
  • Chapter 6?6.37, 6.63
  • Chapter 7?7.1, 7.13, 7.37
  • Quiz Friday on Chapter 6 and 7.1
  • Final Project on Q drive (due May 4th)
Write a Comment
User Comments (0)
About PowerShow.com