sometimes it is helpful to show how different regression models will fit the SAME X,Y scatterplot ...
an example might be useful for your students, etc.
here are some data from the bob hope golf tournament ... showing the relationship between total # of strokes (over the 5 rounds) and tournament earnings
-
900000+
- *
Earnings-
-
-
600000+
-
- *
-
-
300000+
- 2
- * 2
- 4 4
- 5 7 4
0+ 5 6 8 5 * 4 3 2 5 * *
--+---------+---------+---------+---------+---------+----Total strokes
330.0 335.0 340.0 345.0 350.0 355.0
clearly, linear this is NOT
now, a linear model ... quadratic model and a cubic model are all fit using minitab ... the graphs can be seen at:
http://roberts.ed.psu.edu/users/droberts/400406/linear1.jpg
http://roberts.ed.psu.edu/users/droberts/400406/quadratic1.jpg
http://roberts.ed.psu.edu/users/droberts/400406/cubic1.jpg
you MAY have to make the window a bit larger to more clearly see the stats at the top ... like the actual equations
make sure to compare the r squared values ...
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
