s-news
[Top] [All Lists]

question about ms

To: s-news@lists.biostat.wustl.edu
Subject: question about ms
From: Ad Feelders <ad@cs.uu.nl>
Date: Fri, 31 May 2002 14:10:23 +0200
Dear list,

I'm working with Splus6 under windows, and am struggling with ms.

I'm making a call to ms with the first argument minus the loglikelihood of a row in the dataframe "mydata".
"mydata" contains numeric colums named y1,y2 and x.
b1,b2 and r are the parameters of the function to be optimized.
The way I read the documentation, ms should now minimize the sum of the formula provided in the first
argument, evaluated over the rows of mydata.

But I keep getting this error message and I can't figure out why.

Anybod got a clue?

Thanks for your help.

-Ad Feelders

> mydata
  y1 y2 x
1  1  1 2
2  1  0 3
3  0  0 5
4  0  1 4


> ms(~ - (y1 * y2 * log(pmvnorm(c(b1 * x, b2 * x), rho = r)) + y1 * (1 - y2) * log(pnorm(b1 * x) - pmvnorm(c(b1 * x, b2 * x), rho = r)) +
(1 - y1) * log(pnorm( - b1 * x))),data=mydata,start=list(b1=.5,b2=.2,r=.1))

Problem in pmvnorm(c(b1 * x, b2 * x), rho = r): d must be 2 if rho is used, while calling subroutine setup_min


---------------------------------------------------------------------------
Ad Feelders
Universiteit Utrecht
Institute of Information & Computing Sciences
PO Box 80089
3508 TB Utrecht
The Netherlands
ad@cs.uu.nl
Phone: +31 30 2533176


<Prev in Thread] Current Thread [Next in Thread>
  • question about ms, Ad Feelders <=