Download as xlsx, pdf, or txt
Download as xlsx, pdf, or txt
You are on page 1of 12

In this homework, we do Montgomery Problem 6-1 on page 323. We do it using Excel.

Thermal conductivity Product density


y x
0.0480 0.1750
0.0525 0.2200
0.0540 0.2250
0.0535 0.2260
0.0570 0.2500
0.0610 0.2765

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.992827367 Coefficient of determination is R^2
R Square 0.985706181 f. R^2 is 98.6%. This means that 98.6% of the variability in thermal conductivity can be e
Adjusted R Square 0.982132726
Standard Error 0.000585197
Observations 6

ANOVA
df SS MS F Significance F g
Regression 1 9.4463509E-05 9.44635089943E-05 275.841234 7.6985490628357E-05 is
in
Residual 4 1.36982434E-06 3.42456084749E-07 d
Total 5 9.58333333E-05

Coefficients Standard Error t Stat P-value Lower 95%


Intercept 0.024933976 0.00178619168 13.9592946776994 0.000152751 0.0199747128092274
x 0.128521781 0.00773832813 16.6084687421962 7.698549E-05 0.107036737672232

a: Regression coefficients: intercept is 0.0249; slope is 0.1285.


Line equation: Y-hat = 0.0249 + 0.129 x
(I can't get Excel to do a little hat over the y to indicate it's an estimate, so I just write y-hat here.)
RESIDUAL OUTPUT PROBABILITY OUTPUT

Observation Predicted y Residuals Standard Residuals Percentile


1 0.047425288 0.00057471239 1.09800198386899 8.33333333333333
2 0.053208768 -0.0007087678 -1.35411800042809 25
3 0.053851377 0.00014862335 0.283948510654554 41.6666666666667
4 0.053979898 -0.0004798984 -0.91685761682144 58.3333333333333
5 0.057064421 -6.4421178E-05 -0.12307822595264 75
6 0.060470248 0.00052975163 1.01210334867855 91.6666666666667
b. Residuals are here.
x Residual Plot
0.001
DO NOT CHECK 0.0005
BOX FOR

Residuals
0
CONSTANT IS 0 -0.0005600 00 00 00 00 00
1 .18 20 22 24 26 28
0 . 0 0 . 0 . 0 . 0 . 0.
-0.001
x

x Line Fit Plot


0.0800
0.0600
mal conductivity can be explained by the fitted regression model.
0.0400
h. p-value for regression overall is

y
0.0200
called "Sig F" here and is is < 0.05, so
we can reject H0. Regression is 0.0000
significant. Hooray for us. 0.1500 0.2000 0.2500 0.3000
x

g. p-value for intercept


is < 0.05, so we say the
intercept is significantly
different from 0

Normal Probability P
Upper 95% Lower 95.0% Upper 95.0% 0.07
0.06
0.029893239 0.0199747128 0.0298932391
0.05
0.150006824 0.1070367377 0.1500068242 0.04
0.03
y

g. p-value for x is < 0.05, 0.02


so we say the slope is 0.01
significantly different 0
write y-hat here.) from 0 0 10 20 30 40 50 60 7
Sample Percentile

y
0.048
0.0525
0.0535
0.054
0.057
0.061
Residual Plot

00 00 00 00 00 00
20 22 24 26 28 30
0. 0. 0. 0. 0. 0.

x Line Fit Plot

y
Predicted y

0.2000 0.2500 0.3000


x

mal Probability Plot

0 20 30 40 50 60 70 80 90 100
Sample Percentile
Thermal conductivity Product density
y x
0.0480 0.1750
0.0525 0.2200
0.0540 0.2250
0.0535 0.2260
0.0570 0.2500
0.0610 0.2765

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.992827367 k. Sample correlation coefficient is 0.993. We piggyback on our conclusions from parts g
R Square 0.985706181
Adjusted R Square 0.982132726
Standard Error 0.000585197
Observations 6

ANOVA
df SS MS F Significance F
Regression 1 9.4463509E-05 9.44635089943E-05 275.841234 7.6985490628357E-05
Residual 4 1.36982434E-06 3.42456084749E-07
Total 5 9.58333333E-05

Coefficients Standard Error t Stat P-value Lower 95%


Intercept 0.024933976 0.00178619168 13.9592946776994 0.000152751 0.0199747128092274
x 0.128521781 0.00773832813 16.6084687421962 7.698549E-05 0.107036737672232

RESIDUAL OUTPUT PROBABILITY OUTPUT

Observation Predicted y Residuals Standard Residuals Percentile


1 0.047425288 0.00057471239 1.09800198386899 8.33333333333333
2 0.053208768 -0.0007087678 -1.35411800042809 25
3 0.053851377 0.00014862335 0.283948510654554 41.6666666666667
4 0.053979898 -0.0004798984 -0.91685761682144 58.3333333333333
5 0.057064421 -6.4421178E-05 -0.12307822595264 75
6 0.060470248 0.00052975163 1.01210334867855 91.6666666666667
j. Model adequacy check includes our standardized residuals.
Want to make sure not too many outside of +/- 3.
Here's where outliers would show up as well.
x Residual Plot
0.001
0.0005

Residuals
0
-0.0005600 00 00 00 00 00 00 00
.1 .18 .20 .22 .24 .26 .28 .30
0 0 0 0 0 0 0 0
-0.001
x

x Line Fit Plot


conclusions from parts g and h to say this is significant. 0.0800
0.0600
0.0400 y
Predicted y
y

0.0200
0.0000
0.1500 0.2000 0.2500 0.3000
x

i. 95% CI on
intercept is
0.02 - 0.03.
Normal Probability Plot
0.07
0.06
Upper 95% Lower 95.0% Upper 95.0% 0.05
0.029893239 0.0199747128 0.0298932391 0.04
0.150006824 0.1070367377 0.1500068242 0.03
y

0.02
0.01
0
i. 95% CI on 0 10 20 30 40 50 60 70 80 90 100
slope is
0.107 - 0.15 Sample Percentile

y
0.048
0.0525
0.0535
0.054
0.057
0.061
zed residuals.
j: Model adequacy check includes residuals. They look good -
no obvious patterns in them. See MRH p. 320, Figure 6-12 for more residuals.

8 00 00
.2 30
0.

y
Predicted y

ot j: Model adequacy check includes normal probability plot. It looks good -


Data set is too small to really see much, but want a straight line.
See MRH p. 321, Figure 6-13 for more details.

80 90 100

You might also like