[R] Manually setting coefficients in an lm.

[R. Michael Weylandt <michael.weylandt@gmail.com>]

On Nov 12, 2013, at 11:59, <COLLINL at pitt.edu> wrote:

>> Think you're going to have to explicitly define "hand tailor." I
>> haven't a clue what you mean. (Someone else might of course).
> 
> Ah, I want to set a specific coefficient value for each of the terms
> rather than rely on training.  Thus given:
> 
>   y ~ x0 + x1 + x2 + a
> 
> I would like to set:
> 
>    a = 10
>    b0 = 20
>    b2 = 90,
>    etc.
> 
> I'm trying out my own training model and I would like to use the machinery
> of the lm for evaluation and reporting but I would like to fix the beta
> coefficients and other trained values independently of least-squares
> regression.
> 

I'm quite confused: if you set all the coefficients in advance, in what sense are you doing a linear regression? 

Even if this is possible, won't all the other estimates (i.e., standard error of betas) produced be junk since they aren't derived from the associated estimators?

Michael

>    Collin.
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.