comp.lang.idl-pvwave archive: archive » power law fitting for errors on both coordinates

Home » Public Forums » archive » power law fitting for errors on both coordinates

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

power law fitting for errors on both coordinates [message #48387]

Mon, 10 April 2006 19:14

jiajianshen
Messages: 4
Registered: June 2005

Junior Member

I have a data set, which should be a power law, y=a*x^b, both x, y has
their own error measurements. In astronomy, one usually do the log base
and then do a linear fitting to the data (FITEXY).

My advisor hope me do the fitting in the power, not the log based
linear relation. My question is that will the result by the power law
fitting significantly different from the log based linear fitting? The
error is symmetric in normal coordinate, but it will be unsymmetric in
log base, and there is error propagation problem when doing log base.

Do you have any suggestion to do a power law fitting for both x and y
coordinates have their own error measurements? Any information is
highly appreciated.

Thanks,

Jiajian

Report message to a moderator

Previous Topic:	Re: Free source code diagramming programs
Next Topic:	Re: ION script - true colour

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Oct 08 11:45:30 PDT 2025

Total time taken to generate the page: 0.00560 seconds