Hi,

Sometimes the initial 'guess' of your parameters has

to be reasonably close to the right answer in order

for the fit to converge properly. Try this...

b = -0.08

a = 400

f(x) = a*exp(b*x)

fit f(x) 'temp.dat' via a,b

plot 'temp.dat', f(x)

where your data is in the file 'temp.dat' and the

resultant parameter values are

a = 253.968

b = -0.0581577

Hope this helps,

Gavin.

> I'm trying to use the "fit" function on some data

> that I collected in a

> school experiment, but "fit" keeps on dying and

> giving the error "undefined

> value during function evaluation". I am to fit data

> onto the following data:

> 20 78\n25 61\n30 45\n35 34\n40 24\n45 18\n50 13\n55

> 10

> 60 8\n65 6\n70 4\n75 3\n80 2

> Using the function to be fitted as:

> f(x) = a * exp(-1 * b * x) ; Where a, b are the

> constants fit is supposed to

> give me

> I am using gnuplot v. 3.7 patchlevel 1 on a WinXP

> box. After fit dies, I

> used the command "show vars" to see what a and b

> during fitting and they

> were:

> pi = 3.14159265358979

> a = 3.87069662567025e+40

> b = 8.84029163816757e+42

> Is fit dying because a and b are overflowing? How do

> I get a exponential

> decay model out of my data set? I really need the

> values for my report, you

> see.

> I'm not a member of the mailing list, so send the

> suggestions to

> Thanks in advance.

Do you Yahoo!?

Yahoo! SiteBuilder - Free, easy-to-use web site design software

http://sitebuilder.yahoo.com

- -

[[[[ to unsubscribe from info-gnuplot, send an email with body

unsubscribe info-gnuplot

]]]]