Excel 2003 multiple regression coefficients inaccurate

Z

zork

Hi, I am using Excel 2003 and I have found that its multiple regression (via
Data analysis) is very wrong. For instance when I calculate my multiple
regression I get the results:

I am using the model: y = a + b t + c t^2 + d cos(f(t)) + e sin(f(t))

When I use Excel 2003 I get the equation:

Y = 589.9126 + 2.384934t - 0.00389 t^2 + 0 cos + 42.11761 sin

But when I use MINITAB I get the result:

Y = 589 + 2.38 t - 0.00379 t^2 - 64.2 cos + 42.1 sin

Now I plotted both results and Excel 2003 result is a bad fit as it assigns
0 to the cos coefficient. Why does it do this? The Minitab result is a great
fit. Why does Excel 2003 reject this coefficient?

I know the following thread talks about missing coefficients:

http://groups.google.com.au/groups?...c4864f%2439201a00%247d02280a%40phx.gbl&rnum=1

Excel 2003 doesnt seem to give accurate results for this simple multiple
regression problem.

Any help appreciated.
z
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top