Re: help in converting PV-Wave code to IDL

[rkj(at)dukebar.crml.uab.edu (R. Kyle Justice)]

Peter Thorne (peter.thorne@uea.ac.uk) wrote:
: Hi all,

: I was wondering if I could pick your collective brains on this one. I've
: been sent some PV-Wave code which I need to convert to IDL. I've managed
: nearly all of it, in fact there is but one sticking point ...

: In their code they make the following call:

: alpha=multiregress(evects,obs, /no, res=res)

: without having recourse to the PV-Wave documentation I can work out that
: evects is X and obs Y in this multi-linear regression, and I'm guessing
: that res is the residuals which would make sense in terms of further
: statements afterwards. However, I have no idea what keyword no performs
: in the PV-Wave function. The closest equivalent IDL function which I can
: find is regress, but this doesn't seem to have the ability to return the
: residuals on keyword call. I may be barking up the wrong tree though on
: this one.

: Thanks for your help

: Peter

According to the IMSL Stats Reference, /No refers to /No_Intercept.

"When No_INtercept is specified, the total weighted sum of squares is the
sum of squares of yi, the so-called uncorrected total weighted sum of
squares."

Kyle J.