Date:         Fri, 16 May 2008 09:04:56 -0700
Reply-To:     "Huang, Ya" <Ya.Huang@AMYLIN.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Huang, Ya" <Ya.Huang@AMYLIN.COM>
Subject:      Re: Proc Nlin to fit 5 PL, how much the initial value matters ?
Comments: To: Wensui Liu <liuwensui@gmail.com>
In-Reply-To:  <1115a2b00805160902r35ca4f9bv79e56efa1cd65fed@mail.gmail.com>
Content-Type: text/plain; charset="us-ascii"

Thanks Wensui, # of cases means number of data points?

-----Original Message-----
From: Wensui Liu [mailto:liuwensui@gmail.com]
Sent: Friday, May 16, 2008 9:02 AM
To: Huang, Ya
Cc: SAS-L@listserv.uga.edu
Subject: Re: Proc Nlin to fit 5 PL, how much the initial value matters ?

hi, ya,
initial values of para. are important. but in your case, it seems you
have too many # of para. relative to # of cases.


> proc nlin data=xx outest=outparmsfull;  parms a=7 b=-1.25 c=7 d=93
> g=1;
>     model y=d +( (a-d)/( (1+(x/c)**b)**g) ); run;
>
===============================
WenSui Liu
Call for Donations for the China Earthquake!
Blog : statcompute.spaces.live.com
===============================