In: Statistics and Probability
In the production of printed circuit boards, errors in the alignment of electrical connections are a source of scrap. The accompanying table shows the registration error and the temperature used in the production of circuit boards in an experiment in which higher cost material was used.
Registration Error |
Temperature |
||
---|---|---|---|
−10.7 |
278 |
||
3.7 |
320 |
||
3.2 |
299 |
||
10.6 |
321 |
||
−13.8 |
271 |
||
6.2 |
293 |
||
13.7 |
324 |
A. Fit a quadratic regression model and state the quadratic regression.
B. Perform a residual analysis on the results and determine if the regression model is valid.
C. At the 0.05 level of significance, is there a significant quadratic relationship between temperature and registration error?
D. At the 0.05 level of significance, determine whether the quadratic model is a better fit than the linear model.
E. Interpret the meaning of the coefficient of multiple determination.
F. Compute the adjusted r2.
(a) The quadratic regression model is:
y = -960.6411 + 6.0298*x - 0.0094*x²
(b) The residual plot is:
Since a random pattern of the residuals can be seen, the regression model is valid.
(c) The hypothesis being tested is:
H0: β1 = β2 = 0
H1: At least one βi ≠ 0
The p-value is 0.0155.
Since the p-value (0.0155) is less than the significance level (0.05), we can reject the null hypothesis.
Therefore, we can conclude that there is a significant quadratic relationship between temperature and registration error.
(d) R² for the quadratic model is 0.876 while the R² for the linear model is 0.807. Since R² for the quadratic model is morel, it is a better fit than the linear model.
(e) 87.6% of the variation in the model is explained.
(f) Adjusted R² = 0.813
R² | 0.876 | |||||
Adjusted R² | 0.813 | |||||
R | 0.936 | |||||
Std. Error | 4.474 | |||||
n | 7 | |||||
k | 2 | |||||
Dep. Var. | y | |||||
ANOVA table | ||||||
Source | SS | df | MS | F | p-value | |
Regression | 563.5158 | 2 | 281.7579 | 14.08 | .0155 | |
Residual | 80.0614 | 4 | 20.0153 | |||
Total | 643.5771 | 6 | ||||
Regression output | confidence interval | |||||
variables | coefficients | std. error | t (df=4) | p-value | 95% lower | 95% upper |
Intercept | -960.6411 | |||||
x | 6.0298 | 3.7799 | 1.595 | .1859 | -4.4650 | 16.5246 |
x² | -0.0094 | 0.0063 | -1.481 | .2126 | -0.0269 | 0.0082 |