Re: Using NonlinearFit/Regress?
- To: mathgroup at smc.vnet.net
- Subject: [mg33901] Re: Using NonlinearFit/Regress?
- From: shubi at nusun.jinr.ru (Nodar Shubitidze)
- Date: Mon, 22 Apr 2002 00:57:34 -0400 (EDT)
- References: <a9r3gl$lro$1@smc.vnet.net>
- Sender: owner-wri-mathgroup at wolfram.com
You may change the variables:
a=1/2*(Sin[alfa]+1)
b=1/4*(Sin[alfa]+1)*(Sin[betta]+1)
and find "alfa" and "betta" and then find "a" and "b".
All your conditions are fulfillment.
Regards,
Nodar Shubitidze
redrooz at yahoo.com wrote in message news:<a9r3gl$lro$1 at smc.vnet.net>...
> I am trying to estimate the parameters {a,b} in the
> function: x / (1 + a (x-1) + a b x (x-1)) using
> NonlinearFit[...].
>
> The model is known to be a CONCAVE function with
> 0 < a,b < 1 and b < a. NonlinearFit/Regress
> always produces negative parameters (a discontinuous
> function); even if I try to condition them to be
> positive e.g., {a, 0, 1} and {b, 0, 1}.
>
> Any suggestions about using NonlinearFit[...] with the
> above function?
>
> --njg