L PROTECTED]> wrote:
From: Alberto Monteiro <[EMAIL PROTECTED]>
Subject: [R] A doubt about "lm" and the meaning of its summary
To: r-help@r-project.org
Received: Tuesday, 19 August, 2008, 4:31 AM
I have a conceptual problem (sort of). Maybe there's a
simple solution,
maybe not
Hi Alberto,
Please disregard my previous note - I probably had a black-out!!!
--- On Tue, 19/8/08, Alberto Monteiro <[EMAIL PROTECTED]> wrote:
> From: Alberto Monteiro <[EMAIL PROTECTED]>
> Subject: [R] A doubt about "lm" and the meaning of its summary
> To:
Hi Alberto,
In your second case the linear model
y = a*x + b + error
does not hold.
--- On Tue, 19/8/08, Alberto Monteiro <[EMAIL PROTECTED]> wrote:
> From: Alberto Monteiro <[EMAIL PROTECTED]>
> Subject: [R] A doubt about "lm" and the meaning of its summar
I have a conceptual problem (sort of). Maybe there's a simple solution,
maybe not.
First, let me explain the test case that goes ok.
Let x be a (fixed) n-dimensional vector. I simulate a lot of linear models,
y = m x + c + error, then I do a lot of regressions. As expected, the
estimated value
4 matches
Mail list logo
