|
Part 3: The Bottom Line
by Aubrey Kagan
Start ý Calculating
The Odds ý Strange Occurrences ý Can
You Relate? ý Nothing Up My Sleeve
ý Source and PDF
NOTHING UP MY SLEEVE
Now, Iýll show you a sneaky trick. Excel
bases its results on the data without checking the relationship between
them. So, if you create a second column of data with the square of
the PWM and then run a regression for two variables as shown in Photo
6, you get the results shown in Photo 7 (see Example9c.xls
(download
Excel files ).
|

(Click
here to enlarge)
|
Photo 6ýIn this
screen shot, you see the multivariable regression. Note the block
that includes both independent variables PWM and PWM_sq. |
As with other things, one way to prove
a theory is to start out with the answer and see if the model fits.
Discussion of the goodness of the fit are beyond the scope of this
article, but itýs possible to get a direct visual representation by
using the Line Fit Plots option in the Residuals section of the Regression
window. The output options also should be in a new workbook.
TO SUM IT UP
In any software application, most of
us use only 20% of the features. It seems a shame to waste the resources
available to us, especially because they could significantly affect
easing the effort of creating a new design. I hope this series has
provided you some insight as to the features available and how to
use them.
PREVIOUS
NEXT
Circuit Cellar provides up-to-date information for engineers. Visit
www.circuitcellar.com for
more information and additional articles.
For subscription information, call (860) 875-2199, subscribe@circuitcellar.com
or subscribe online.
ýCircuit Cellar, the Magazine for Computer Applications. Posted with
permission. |