ASE Home Page Products Download Purchase Support About ASE
ChartDirector General
Forum HomeForum Home   SearchSearch

Message ListMessage List     Post MessagePost Message

  Exponential and Logarithmic Conditions Compared to excel
Posted by Scott on Mar-30-2009 02:12
Attachments:
Hi,

I've compared some of the regressions in Chart director to those in Excel.

Both Exponential and logarithmic regressions appear to be working however they do not generate identical results as Excel.

Please review the attached dataset. CD gives a regression of
4.66*ln(x)-0.82 and an R^2 of 0.9468. Excel has a different equation
and a higher R^2 indicating a more accurate regression. Both
regressions should generate the same results shouldn't they.
CD_Fail.xls
CD_Fail.xls

25.90 Kb

  Re: Exponential and Logarithmic Conditions Compared to excel
Posted by Peter Kwan on Mar-31-2009 01:12
Attachments:
Hi Scott,

I have just checked myself. Using your data in Excel 2007, with logarithmic regression, I get the formula:

y = 4.549 * ln(x) - 0.359
r^2 = 0.955

I tried the same data in ChartDirector by hard coding them in the source code. I have attached my testing code (which is written in ASP/VBScript), and a screen shot to show the result. It is exactly the same as Excel.

If in your case, the result is different, is it possible to attach your test code and the resulting image?

Hope this can help.

Regards
Peter Kwan
paramcurve.png
paramcurve.asp
paramcurve.asp

3.43 Kb

  Re: Exponential and Logarithmic Conditions Compared to excel
Posted by Scott on Mar-31-2009 03:34
I've rechecked my analysis and realize that you are right. I do not know how I got it into my head that the two regressions are different.

I apologize for the error.