Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: More on nonlinear models
Replies: 2   Last Post: Oct 31, 1996 6:16 PM

 Messages: [ Previous | Next ]
 James Lang Posts: 48 Registered: 12/6/04
More on nonlinear models
Posted: Oct 30, 1996 1:22 AM

One comment that Dick Scheaffer made was

>It should be pointed out to students that fitting a least squares line
>to the transformed data is not the same as fitting a nonlinear
>model directly by least squares.

If you would like an example, I have some data on Toyota Corolla prices
that I got from the Orlando Sentinel classified ads. The variables are the
age of the car and the advertised price.

Age Price
1 11500
1 12000
3 8000
5 6000
6 3500
6 3000
6 4500
6 4000
8 4000
9 2200
10 3000
12 1000
12 600
16 600

If you do a ExpReg you get y = 14837.73(0.80851^x) which has SSE = 6431130.

Using a couple data points, (1,11500) and (16,600), that appeared to be a
good visual fit to estimate a,b
I got y = 14002.35(0.82129^x) which has SSE = 6302004.

The point is that the visual fit gave a smaller SSE than the regression on
this with any data set that "appears" to be exponential but has some
deviation from the exponential model.

James Lang
Valencia Community College
Orlando, FL

Date Subject Author
10/30/96 James Lang
10/30/96 Joe H Ward
10/31/96 Paul Velleman