Title | CH5 Quadratic Regression notes |
---|---|
Author | Andrea Billings |
Course | Microeconomics |
Institution | Fashion Institute of Technology |
Pages | 7 |
File Size | 368.1 KB |
File Type | |
Total Downloads | 21 |
Total Views | 161 |
PPT...
Chapter 15
15-1
Statistics for Managers Using Microsoft Excel
Chapter 5 Multiple Regression Model Building
Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Chap 15-1
Nonlinear Relationships DCOVA
The relationship between the dependent variable and an independent variable may not be linear Can review the scatter plot to check for nonlinear relationships Example: Quadratic model
Yi β0 β1X1i β2 X21i εi
The second independent variable is the square of the first variable
Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Basic Business Statistics, 10/e
Chap 15-2
© 2006 Prentice Hall, Inc.
Chapter 15
15-2
Quadratic Regression Model DCOVA Model form:
Yi β0 β1X1i β2 X21i εi
where: β0 = Y intercept β1 = regression coefficient for linear effect of X on Y β2 = regression coefficient for quadratic effect on Y εi = random error in Y for observation i
Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Chap 15-3
Linear vs. Nonlinear Fit DCOVA
Y
Y
X
X Linear fit does not give random residuals
residuals
residuals
X
X
Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Basic Business Statistics, 10/e
Nonlinear fit gives random residuals Chap 15-4
© 2006 Prentice Hall, Inc.
Chapter 15
15-3
Quadratic Regression Model DCOVA
Yi β0 β1X1i β2 X21i εi
Quadratic models may be considered when the scatter plot takes on one of the following shapes: Y
Y
β1 < 0
X1
Y
β1 > 0
β2 > 0
β2 > 0
X1
Y
β1 < 0
X1
β2 < 0
β1 > 0
X1
β2 < 0
β1 = the coefficient of the linear term β2 = the coefficient of the squared term Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Chap 15-5
Testing the Overall Quadratic Model DCOVA
Estimate the quadratic model to obtain the regression equation: ˆ i b0 b1X1i b2 X1i2 Y
Test for Overall Relationship H0: β1 = β2 = 0 (no overall relationship between X and Y) H1: β1 and/or β2 ≠ 0 (there is a relationship between X and Y)
FSTAT =
MSR MSE
Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Basic Business Statistics, 10/e
Chap 15-6
© 2006 Prentice Hall, Inc.
Chapter 15
15-4
Testing for Significance: Quadratic Effect DCOVA
Testing the Quadratic Effect
Compare quadratic regression equation
Yi b0 b1X1i b2 X1i2 with the linear regression equation
Yi b0 b1X1i
Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Chap 15-7
Testing for Significance: Quadratic Effect (continued)
Testing the Quadratic Effect
DCOVA
Consider the quadratic regression equation
Yi b0 b1X1i b2 X1i2
Hypotheses H0: β2 = 0 (The quadratic term does not improve the model) H1: β2 0 (The quadratic term improves the model)
Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Basic Business Statistics, 10/e
Chap 15-8
© 2006 Prentice Hall, Inc.
Chapter 15
15-5
Testing for Significance: Quadratic Effect (continued)
Testing the Quadratic Effect
DCOVA
Hypotheses H0: β2 = 0 (The quadratic term does not improve the model) H1: β2 0 (The quadratic term improves the model)
The test statistic is
b β2 t STAT 2 Sb 2
d.f. n 3
where: b2 = squared term slope coefficient β2 = hypothesized slope (zero) Sb = standard error of the slope 2
Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Chap 15-9
Testing for Significance: Quadratic Effect (continued)
Testing the Quadratic Effect
DCOVA
Compare r2 from simple regression to adjusted r2 from the quadratic model
If adj. r2 from the quadratic model is larger than the r2 from the simple model, then the quadratic model is likely a better model
Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Basic Business Statistics, 10/e
Chap 15-10
© 2006 Prentice Hall, Inc.
Chapter 15
15-6
Example: Quadratic Model DCOVA 3
1
7
2
8
3
15
5
22
7
33
8
40
10
54
12
67
13
70
14
78
15
85
15
87
16
99
17
Purity increases as filter time increases: Purity vs. Time 100 80 60
Purity
Purity
Filter Time
40 20 0 0
5
10
15
20
Time
Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Chap 15-11
Example: Quadratic Model (continued)
Simple regression results:
DCOVA
Y^ = -11.283 + 5.985 Time Coefficients Intercept Time
Standard Error
t Stat
t statistic, F statistic, and r2 are all high, but the residuals are not random:
P-value
-11.28267
3.46805
-3.25332
0.00691
5.98520
0.30966
19.32819
2.078E-10
Regression Statistics 0.96888
Adjusted R Square
0.96628
Standard Error
6.15997
F 373.57904
Time Residual Plot
Significance F
10
2.0778E-10 Residuals
R Square
5 0 -5 0
5
10
15
20
-10 Time Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Basic Business Statistics, 10/e
Chap 15-12
© 2006 Prentice Hall, Inc.
Chapter 15
15-7
Example: Quadratic Model in Excel (continued)
Quadratic regression results:
DCOVA
Y^ = 1.5387 + 1.5650 Time + 0.2452 (Time)2 Coefficients
Standard Error
Intercept
1.5387
2.2447
0.6855
0.5072
Time
1.5650
0.6018
2.6005
0.0247
Time-squared
0.2452
0.0326
7.5241
0.0000
Time Residual Plot
P-value
10 Resi duals
t Stat
5 0 -5
F
R Square
0.9949
Adjusted R Square
0.9940
Standard Error
2.5951
1080.7330
5
Significance F 2.368E-13
The quadratic term is significant and improves the model: adj. r2 is higher and SYX is lower, residuals are now random
10
15
20
Time-squared Residual Plot 10
Statistics for Managers Using Microsoft Excel® 7e Copyright ©2014 Pearson Education, Inc.
Basic Business Statistics, 10/e
0
Time
Resi duals
Regression Statistics
5 0 -5
0
100
200
300
400
Time-squared
Chap 15-13
© 2006 Prentice Hall, Inc....