On 17 Sep 2014, at 05:58 , isabe...@ghement.ca wrote:
>
> The example is based on a real data set (please keep it
> confidential), which is attached to this e-mail as mod.data.csv.
"Don't worry, it will stay between you two and the Internet"
(Actually, the list software seems to ha
What's going on is very simple. I apologize for not getting my thoughts
organized from the start.
Confidence bands for mu_y (the expected value of y) will have straight
line edges if and only if there is a *single predictor* in the model.
If there are two or more predictors the edges will
Hi Rolf,
BODY { font-family:Arial, Helvetica, sans-serif;font-size:12px; }
Thanks very much for your response. You are right - my simulated
example works as intended, so it can't be used to get to the bottom of
this problem (if it is a problem at all).
Here is
When I run your code (in the single predictor case) I get exactly what I
would expect. In particular the standard errors are indeed proportional
to the (absolute) value of x, and the standard error is indeed 0 at x = 0.
The proportionality constant is exactly what it should be, explicitly
s/
Hi everyone,
It appears my R code didn't come through the first time (thanks for letting me
know, Ista). Here is my message again:
Could there be an error in the predict() function in R for models without
intercepts which include one or more predictors?
When using the predict() function to
5 matches
Mail list logo