s-news
[Top] [All Lists]

[S] use of ms() on complicated functions

To: <s-news@wubios.wustl.edu>
Subject: [S] use of ms() on complicated functions
From: "Peter Holck" <PHolck@anhb.org>
Date: Tue, 26 May 1998 10:39:16 -0800
Reply-to: "Peter Holck" <PHolck@anhb.org>
Sender: owner-s-news@wubios.wustl.edu
I'm trying to use ms() to select parameters of a function such that the intraclass correlation of a pair of vectors transformed by this function is maximized:  Specifically, I want to find a logistic transformation for a pair of vectors with parameters of the logistic function selected to maximize the intraclass correlation of the transformed vectors (or equivalently, minimize 1/r or -r in order to make use of ms(), which minimizes rather than maximizes).
 
I understand that utilizing derivatives and hessians of a function (using deriv3(), for example) should improve the performance (and stability?) of the ms() routine.  However, the function I am using becomes rather complicated (see below), and I'm not certain if deriv3() would be applicable.   My problem is that in simulations without specifying derivative/hessians I often get singular convergence or false convergence messages from ms(), and I am concerned about trusting the resulting function parameters chosen by the ms() routine.  Furthermore, I am surprised that in the occasional simulation no convergence is reached by ms() even after 100 iterations. 
I have set f.tolerance to 0 in the ms.control options.  I am using Splus version 4.3, win95.
 
Any insight into the advisability of using ms() in this manner or into refinements or changes in my approach to this problem would be much appreciated.
 
The specific function goal is:
maximize (x1-x2)/(x1+x2)
    where:
    x1=mean(y1-y2)^2/2,   and
    x2=2*var((y1+y2)/2)
            where:
            y1 = 1/(1 + exp( - (a + b1 * z1 + b2*(z1^2)))) and
            y1 = 1/(1 + exp( - (a + b1 * z2 + b2*(z2^2))))
 
                where z1 and z2 are vectors of equal length (paired)
 
My goal is to find parameters a, b1, and b2 for a particular z1 and z2
 
Thanks in advance -
 
Peter Holck
Alaska Native Health Board
pholck@anhb.org           
<Prev in Thread] Current Thread [Next in Thread>
  • [S] use of ms() on complicated functions, Peter Holck <=