Andrew Eskander
Math 11L
April 18, 2008
Lab 3
1.
The equation of the regression line is
BMR = 157 + 0.187 * Mass
.
50000
40000
30000
20000
10000
0
9000
8000
7000
6000
5000
4000
3000
2000
1000
0
Mass
BMR
Scatterplot of BMR vs Mass
2.
The regression line predicts the BMR of the Eastern Chipmunk to be
173.24
(BMR =
157 + .187(87.4) = 173.34) and the BMR of the Aardvark to be
9139.44
(BMR = 157
+ .187(6144) = 9133).
These predictions are much more than the observed values (90
and 6144 respectively) of the Eastern Chipmunk and Aardvark.
3.
Judging from both plots, I think that my linear regression model is not appropriate for
predicting basal metabolic rate from body size because there are several outliers that
distort the predictions of the BMR.
There are too many points far from the line in the
residual and scatterplots that offset the points clustered in proximity to the line in
each plot, thus, making my model unevenly spread about the line and overall
inappropriate for predictions.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
50000
40000
30000
20000
10000
0
9000
8000
7000
6000
5000
4000
3000
2000
1000
0
Mass
BMR
Scatterplot of BMR vs Mass
50000
40000
30000
20000
10000
0
5000
2500
0
2500
5000
Mass
Residual
Residuals Versus Mass
(response is BMR)
4.
Taking logarithms of the variables reduces the skewness, however the histograms of
the two new transformed variables are still somewhat skewed to the right, partly
because many of the mammals have a low mass.
This is the end of the preview.
Sign up
to
access the rest of the document.
 Fall '08
 Schweinsberg
 Math, Statistics, Econometrics, Forecasting, Linear Regression, Probability, Regression Analysis, Errors and residuals in statistics, BMR

Click to edit the document details