In: Statistics and Probability
Please Use Minitab
The provided data is taken from an experiment aimed at applying the two factors of temperature and alloy type as predictors of battery life.
RandOrder | Alloy | Temperature | Life |
9 | A1 | 5 | 140 |
6 | A1 | 5 | 84 |
1 | A1 | 5 | 165 |
4 | A1 | 5 | 180 |
7 | A1 | 70 | 34 |
10 | A1 | 70 | 80 |
2 | A1 | 70 | 40 |
5 | A1 | 70 | 75 |
3 | A1 | 135 | 15 |
12 | A1 | 135 | 72 |
8 | A1 | 135 | 65 |
11 | A1 | 135 | 53 |
19 | A2 | 5 | 160 |
15 | A2 | 5 | 169 |
17 | A2 | 5 | 198 |
20 | A2 | 5 | 136 |
21 | A2 | 70 | 136 |
16 | A2 | 70 | 106 |
24 | A2 | 70 | 122 |
18 | A2 | 70 | 115 |
22 | A2 | 135 | 20 |
14 | A2 | 135 | 53 |
13 | A2 | 135 | 65 |
23 | A2 | 135 | 40 |
27 | A3 | 5 | 143 |
26 | A3 | 5 | 173 |
34 | A3 | 5 | 115 |
32 | A3 | 5 | 160 |
33 | A3 | 70 | 174 |
29 | A3 | 70 | 150 |
28 | A3 | 70 | 120 |
25 | A3 | 70 | 139 |
36 | A3 | 135 | 91 |
30 | A3 | 135 | 77 |
35 | A3 | 135 | 99 |
31 | A3 | 135 | 55 |
1. Develop a multiple linear regression model based on the data. Comment on the significance of linear regression (α = 0.05) and the adequacy of the model for explaining variation in the response variable.
2. Plot and analyze the residuals from the experiment. Comment on flaws in the model if any, regarding the assumptions of linear regression.
1)
Regression Analysis: Life versus Temperature, Alloy
Method
Categorical predictor coding (1, 0)
Analysis of Variance
Source DF Adj
SS Adj MS F-Value P-Value
Regression 3 62483
20827.8 24.36 0.000
Temperature 1 52080
52080.2 60.91 0.000
Alloy 2
10403 5201.6
6.08 0.006
Error
32 27361 855.0
Lack-of-Fit 5 10328
2065.6 3.27 0.019
Pure Error 27 17033
630.9
Total
35 89845
Model Summary
S R-sq
R-sq(adj) R-sq(pred)
29.2411 69.55%
66.69% 61.34%
Coefficients
Term
Coef SE Coef T-Value P-Value VIF
Constant
133.8 10.6
12.61 0.000
Temperature -0.7167 0.0918
-7.80 0.000 1.00
Alloy
A2
26.4 11.9
2.21 0.034 1.33
A3
41.1 11.9
3.44 0.002 1.33
Regression Equation
Alloy
A1 Life = 133.8 - 0.7167 Temperature
A2 Life = 160.2 - 0.7167 Temperature
A3 Life = 174.8 - 0.7167 Temperature
Fits and Diagnostics for Unusual Observations
Obs Life
Fit Resid Std Resid
27 115.00 171.25 -56.25 -2.06 R
R Large residual