| Maximum likelihood fitting of an exponential fitting function in IDL [message #91591] |
Tue, 04 August 2015 09:01  |
atmospheric physics
Messages: 121
Registered: June 2010
|
Senior Member |
|
|
Dear All,
I have an exponential decay function as defined below:
FUNCTION tfunct,X,E
T = EXP(-((1000.0d*X)^E[1])/E[0])
RETURN,T
END
I was using MPFITFUN to get the fitting coefficients. Though this is working fine, I find that the fitting line is deviating away from the more likelihood points in the case when there were some outliers. Can I know if there is a maximum likelihood fitting function in IDL so that I can get the fitting line close to more number of points?
Look forward to your suggestions,
Thanking you,
With regards,
Madhavan
|
|
|
|
| Re: Maximum likelihood fitting of an exponential fitting function in IDL [message #91592 is a reply to message #91591] |
Tue, 04 August 2015 10:27  |
Bill Nel
Messages: 31
Registered: October 2010
|
Member |
|
|
On Tuesday, August 4, 2015 at 12:01:41 PM UTC-4, Madhavan Bomidi wrote:
> Dear All,
>
> I have an exponential decay function as defined below:
>
> FUNCTION tfunct,X,E
> T = EXP(-((1000.0d*X)^E[1])/E[0])
> RETURN,T
> END
>
> I was using MPFITFUN to get the fitting coefficients. Though this is working fine, I find that the fitting line is deviating away from the more likelihood points in the case when there were some outliers. Can I know if there is a maximum likelihood fitting function in IDL so that I can get the fitting line close to more number of points?
>
> Look forward to your suggestions,
>
> Thanking you,
> With regards,
> Madhavan
You might try taking the log of your data before fitting it (to a line).
--Wayne
|
|
|
|
| Re: Maximum likelihood fitting of an exponential fitting function in IDL [message #91675 is a reply to message #91591] |
Wed, 12 August 2015 10:35 |
Jeremy Bailin
Messages: 618
Registered: April 2008
|
Senior Member |
|
|
On Tuesday, August 4, 2015 at 12:01:41 PM UTC-4, Madhavan Bomidi wrote:
> Dear All,
>
> I have an exponential decay function as defined below:
>
> FUNCTION tfunct,X,E
> T = EXP(-((1000.0d*X)^E[1])/E[0])
> RETURN,T
> END
>
> I was using MPFITFUN to get the fitting coefficients. Though this is working fine, I find that the fitting line is deviating away from the more likelihood points in the case when there were some outliers. Can I know if there is a maximum likelihood fitting function in IDL so that I can get the fitting line close to more number of points?
>
> Look forward to your suggestions,
>
> Thanking you,
> With regards,
> Madhavan
ML_DISTFIT in JBIU does maximum likelihood fitting:
http://www.simulated-galaxies.ua.edu/jbiu/
It's kind of clunky, though, so YMMV.
-Jeremy.
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
