[S] nlregb() and nls()

Joe Mortzheim (jmortz@snake1.cr.usgs.gov)
Wed, 22 Apr 1998 15:07:44 -0400


Please excuse the question if it is overly simple. However, I would appreciate any advice.

I am using nls() and wish to place maximum upper and lower limits on parameter estimates (non-negativity for example). It would seem that nlregb() would fit the bill if I could make sure the syntax is correct.

My nls call looks like this:

# DHT, DBH, T.IN, S.IND, N.BA are columns in the matrix called Temp.Regression.Data

The.Formula <- DHT ~ 4.5+(B1*(1-exp(-2*DBH))^B3)*(T.IN^B4)*(S.IND^B5)*(N.BA^B6)

Output <- nls( The.Formula,
Temp.Regression.Data,
start=list(B1=234, B2=22, B3=0.45, B4=0.22, B5=2.1, B6=3.4),
trace=T)

How would that be translated into an identical call to nlregb() with, for example limits that B1 < 250 and B4 > 0?

If someone could give an example on how to translate it I would appreciate it.

Also, is there a simple, concise description of how to properly call nlregb() and nls() anywhere out there on the web?

Thank you.
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news