R help - Jan 2006 - Breakpoints for multiple variables using Segmented

Hi all,

I am using the package ?Segmented? to estimate logistic regression models
with unknown breakpoints (see Muggeo 2003 Statistics in Medicine
22:3055-3071). In the documentation it suggests that it might be possible to
include several variables with breakpoints in the same model: ?Z = a vector
or a matrix meaning the (continuous) explanatory variable(s) having
segmented relationships with the response?. However, the syntax for
including multiple ?Z? and ?psi? (?starting values for the break-point(s)?)
is not stated. Does anyone have any suggestions?

Here is an example of correct code for detecting single breakpoint:

model.seg<-segmented.glm(obj = model.glm, Z = predictor_variable, psi = 2 ,
it.max = 50)

Thanks very much for your help.


Matthew G. Betts, Ph.D.
NB??Cooperative Fish and Wildlife Research Unit
Faculty of Forestry and Environmental Management
University of New Brunswick??
UNB Tweedale Centre
Hugh John Flemming Forestry Complex
1350 Regent St., Fredericton, N.B.
E3C 2G6
(506) 447-3408
http://www.unb.ca/web/acwern/people/mbetts/mbetts.htm

Dear Matthew,
Currently segmented() performs multiple (say L>1) estimation of 
breakpoints in GLM, namely:

1)L breakpoints for the same variable x, e.g.:
segmented(obj.glm, Z=x, psi=c(psi1,psi2,psi3))

2)L breakpoints for L explanatory `segmented' variables, e.g.:
segmented(obj.glm, Z=cbind(x1,x2,x3), psi=c(psi1,psi2,psi3))
In this case it is assumed that each x has just one breakpoint (psi1 for 
x1, psi2 for x2, and so on). This is, actually , a limit - I know. 
Currently I have no time :-(, but sooner or later I should update the 
package..

Hope this helps,
vito



Matthew Betts wrote:
> Hi all,
> 
> I am using the package Segmented to estimate logistic regression models
> with unknown breakpoints (see Muggeo 2003 Statistics in Medicine
> 22:3055-3071). In the documentation it suggests that it might be possible
to
> include several variables with breakpoints in the same model: Z = a vector
> or a matrix meaning the (continuous) explanatory variable(s) having
> segmented relationships with the response. However, the syntax for
> including multiple Z and psi (starting values for the break-point(s))
> is not stated. Does anyone have any suggestions?
> 
> Here is an example of correct code for detecting single breakpoint:
> 
> model.seg<-segmented.glm(obj = model.glm, Z = predictor_variable, psi =
2 ,
> it.max = 50)
> 
> Thanks very much for your help.
> 
> 
> Matthew G. Betts, Ph.D.
> NB Cooperative Fish and Wildlife Research Unit
> Faculty of Forestry and Environmental Management
> University of New Brunswick 
> UNB Tweedale Centre
> Hugh John Flemming Forestry Complex
> 1350 Regent St., Fredericton, N.B.
> E3C 2G6
> (506) 447-3408
> http://www.unb.ca/web/acwern/people/mbetts/mbetts.htm
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
> 
-- 
===================================Vito M.R. Muggeo
Dip.to Sc Statist e Matem `Vianelli'
Universit di Palermo
viale delle Scienze, edificio 13
90128 Palermo - ITALY
tel: 091 6626240
fax: 091 485726/485612