My Math Forum  

Go Back   My Math Forum > College Math Forum > Applied Math

Applied Math Applied Math Forum

LinkBack Thread Tools Display Modes
February 12th, 2009, 02:11 AM   #1
Joined: Feb 2009

Posts: 1
Thanks: 0

fit one nonlinear function to four datasets

Dear all,

I encountered a math problem when trying to find a set of robust and accurate parameters for a PDF, a nonlinear function with 3 parameters. The problem is:

I have 4 datasets (X1,Y1), (X2,Y2),(X3,Y3),(X4,Y4) for a certain distribution. Each set, (Xi, Yi), contains about 400 data and was obtained from a different observational experiment. So it is assumed that different datasets should have different error variance. I am trying to use nonliner regression to obtain an empirical function from the four data sets. As the variance in each data set is unknown, it is impossible to obtain a weighting coefficient if i put them together. Is there any other solution for this case ?

The empirical function is a*X^b/(1+exp(c*X)).

I will appreciate very much if some could give me some support. Thank you!

yjwang05 is offline  

  My Math Forum > College Math Forum > Applied Math

datasets, fit, function, nonlinear

Thread Tools
Display Modes

Similar Threads
Thread Thread Starter Forum Replies Last Post
Applying proportional relationships between datasets Rincewind Applied Math 3 March 4th, 2014 11:30 AM
Comparing three datasets and two variables at the same time? Spoilerman Algebra 0 July 5th, 2013 07:13 AM
nonlinear DE elimax Calculus 3 July 7th, 2011 02:28 AM
nonlinear regression method_man Applied Math 3 June 28th, 2011 10:20 AM
nonlinear function wannabe1 Linear Algebra 2 February 2nd, 2010 09:56 PM

Copyright © 2019 My Math Forum. All rights reserved.