Logarithmic Curve Fit

Beta releases, announcements and anything else that doesn't fit below. Please read the <B>Welcome</B> topic.

Moderator: DPlotAdmin

Post Reply
SHAWN18
Posts: 1
Joined: Fri May 11, 2007 10:48 am

Logarithmic Curve Fit

Post by SHAWN18 »

I have an linear x, y scatter plot. And I am trying to fit a Logarithmic curve to the data. How is this done?
User avatar
DPlotAdmin
Posts: 2312
Joined: Tue Jun 24, 2003 9:34 pm
Location: Vicksburg, Mississippi
Contact:

Post by DPlotAdmin »

The Polynomial Curve Fit command on the Generate menu will (in most cases) use an equation form that matches the scaling on the axes. For example for linear X, linear Y you'll get

y=C0+C1*x+C2*x^2+....

For linear x, logarithmic y you'll get

log10(y)=C0+C1*x+C2*x^2+....

and for logarithmic x, linear y you'll get

y=C0+C1*log10(x)+C2*(log10(x))^2+...

So if that's the form you want, right-click on the graph and select "Logarithmic X, Linear Y".
Visualize Your Data
support@dplot.com
kapqh4
Posts: 1
Joined: Wed Jan 14, 2009 5:36 pm

Post by kapqh4 »

I have searched these forums but I can't find an answer to this. This thread is as close as I've found.

Is there a way to do a natural log curve (base e) instead of a log base 10?
User avatar
DPlotAdmin
Posts: 2312
Joined: Tue Jun 24, 2003 9:34 pm
Location: Vicksburg, Mississippi
Contact:

Post by DPlotAdmin »

If your plot's scaling is set to Linear X, Logarithmic Y, then Generate>Polynomial Curve Fit gives you both.

For example:

log(y) = C(0) + C(1)*x + C(2)*x^2 + ...
(log() = common logarithm)
Coefficients
C(00) = 0.346914605862
C(01) = -1.64979805157
C(02) = 3.25720861427

or

ln(y) = D(0) + D(1)*x + D(2)*x^2 + ...
(ln() = natural logarithm)
Coefficients
D(00) = 0.7988004
D(01) = -3.7988004
D(02) = 7.5

On the other hand if you're looking for base e scaling - no, sorry, that isn't in there.
Visualize Your Data
support@dplot.com
Post Reply