Re: Troubles in NonlinearFit-Why always failed?

*To*: mathgroup at smc.vnet.net*Subject*: [mg65982] Re: Troubles in NonlinearFit-Why always failed?*From*: "simon yang" <yanshanguke at 163.com>*Date*: Tue, 25 Apr 2006 05:19:19 -0400 (EDT)*References*: <e2fkib$blb$1@smc.vnet.net>*Sender*: owner-wri-mathgroup at wolfram.com

I have found the solutions. The key lies here, the functions Gc[x] and L[x] used in Epsilon2[x] was not be replaced by their definement. If I replace the Gc[x] and L[x] in Epsilon2[x] directly by their definement, then NolinearRegress work. So how to make sure functions were replaced by their definement when I use them in other functions? The NonlinearRegress use too much time though I had set the MaxIterations to 10, about 1 hour, in mathematica! But in other software developed by C language , 10 minutes just be used though the function have hundreds of parameters. How to make NonlinearFit in mathematica more quick?