[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: question about a function to calculate regression

From: Paul Kienzle
Subject: Re: question about a function to calculate regression
Date: Tue, 13 Jun 2006 06:47:34 -0400

On Jun 12, 2006, at 11:56 PM, Tracy Feldman wrote:

To whom it may concern:
I am trying to conduct a simple regression of two vectors (x and y values--regressing y as a function of x).  I had been using the 'regress' function in matlab:
[b, bint, r, rint, stats] = regress(y, x);

where I am particularly interested in:
1. b, which is the vector of parameter estimates for both the slope *and* the intercept (not necessarily zero),
2. stats, which is a vector with the R-squared, F, and P values 
I have been searching through the octave help menu and octave websites, but I have not found anything that seems to do what I want it to do.  I may be misunderstanding things, but the function 'gls' does not seem right, as I cannot figure out what it returns that is like an r-squared or a p-value, and it also seems to require the assumption that the line intersects with the origin.  If I am right, I would appreciate any help people might offer regarding this issue, and suggestions for functions that might fit the bill.  If the function can indeed do what I want it to do, I am not sure how to write the syntax such that it works.

octave-forge has regress:*checkout*/octave/octave-forge/main/ statistics/regress.m?revision=1.3

- Paul

reply via email to

[Prev in Thread] Current Thread [Next in Thread]