| Help with constrained fit in IDL [message #12009] |
Thu, 18 June 1998 00:00  |
lengel
Messages: 1
Registered: June 1998
|
Junior Member |
|
|
Hi all,
I have written some IDL code to fit our data with a power law and it
works fine. Now I am trying to constrain the fit to use a specific value
of an exponent and only vary the factor in front of the expression. I
cannot get this to work properly, and I think it should be simpler
than I am making it out to be. I figure this is due to my relative
inexperience with the FUNCT keyword and how the data fitting is done
in IDL. Can anyone give me a hint for proceeding with performing a fit
to data that constrains one of the factors to a specific value?
Thanks for any pointers that are given.
--
George Lengel mailto:lengel@lps.umd.edu
Laboratory for Physical Sciences OR mailto:lengel@glue.umd.edu
University of Maryland
***I certainly do not speak for LPS or UMd in any capacity******
|
|
|
|
| Re: Help with constrained fit in IDL [message #12120 is a reply to message #12009] |
Tue, 23 June 1998 00:00  |
Heiko H�nnefeld
Messages: 7
Registered: June 1998
|
Junior Member |
|
|
Hi all,
I am sorry for posting the last issue in HTML - format.
maybe some of you could not read the message, so I posted it again - in
plain text this time.
Heiko
> George A. Lengel wrote:
>
>> Hi all,
>>
>> I have written some IDL code to fit our data with a power law and it
>>
>> works fine. Now I am trying to constrain the fit to use a specific
>> value
>> of an exponent and only vary the factor in front of the expression.
>> I
>> cannot get this to work properly, and I think it should be simpler
>> than I am making it out to be. I figure this is due to my relative
>> inexperience with the FUNCT keyword and how the data fitting is done
>>
>> in IDL. Can anyone give me a hint for proceeding with performing a
>> fit
>> to data that constrains one of the factors to a specific value?
>>
>> Thanks for any pointers that are given.
>>
>> --
>> George Lengel mailto:lengel@lps.umd.edu
>> Laboratory for Physical Sciences OR mailto:lengel@glue.umd.edu
>> University of Maryland
>> ***I certainly do not speak for LPS or UMd in any capacity******
>
> Hi George,
>
> we have modified the Curvefit from IDL, so that you can fix paramters
> to a specified value.
> The result is, that you need only one Fitfunction-procedure for your
> fit
> instead of defing a new procedure for every set of free parameters.
> Furthermore we corrected the curvefit-procedure in some points.
> E.g. the errors for the parameters are now the correct ones, also you
> have a correlation matrix as output.
> Send an email to me, if you want to get my curvefit-routine.
>
> Heiko
> --
>
> Heiko H�nnefeld
>
> HASYLAB / DESY
Notkestr. 85
22603 Hamburg
Tel.: 040 / 8998-2698
Fax.: 040 / 8998-2787
> e-mail: hhuenne@desy.de
>
>
>
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
