R help - Jul 2007 - Regression with Missing values. na.action?

Hi all,

Can you please tell me what is the problem here.

My regression eq is  y = B0 + B1X1 +B2X2 +e
And i am interested in coefficient B1

I am doing regression with two cases:

1) reg<-lm(y ~ X1 + X2, sam) where sam is the data

2) reg<-lm(y ~ X1 + X2, sam, na.action= na.exclude) . I have missing values
in X1


but the values of coefficient is not consistent in two cases.

Actually B1 in case one sould be smaller than B1 in case 2. But sometimes it
comes greater.

I can't figure it out. Is there some problem with *na.action ? *My sample
size is 100


*Regards,*
*Vaibhav*

	[[alternative HTML version deleted]]

na.exclude should give the same results as na.omit, which is the
default na.action.  Is the number of complete cases the same in these
two regressions?

On 26/07/07, Vaibhav Gathibandhe <gathibandhe.vaibhav at gmail.com>
wrote:
> Hi all,
>
> Can you please tell me what is the problem here.
>
> My regression eq is  y = B0 + B1X1 +B2X2 +e
> And i am interested in coefficient B1
>
> I am doing regression with two cases:
>
> 1) reg<-lm(y ~ X1 + X2, sam) where sam is the data
>
> 2) reg<-lm(y ~ X1 + X2, sam, na.action= na.exclude) . I have missing
values
> in X1
>
>
> but the values of coefficient is not consistent in two cases.
>
> Actually B1 in case one sould be smaller than B1 in case 2. But sometimes
it
> comes greater.
>
> I can't figure it out. Is there some problem with *na.action ? *My
sample
> size is 100
>
>
> *Regards,*
> *Vaibhav*
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at stat.math.ethz.ch 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.
>

-- 
================================David Barron
Said Business School
University of Oxford
Park End Street
Oxford OX1 1HP