[R] Compare effects between lm-models

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

[R] Compare effects between lm-models

Rense Nieuwenhuis
Dear helpeRs,

I'm estimating a series of linear models (using lm) in which in every  
new model variables are added. I want to test to what degree the new  
variables can explain the effects of the variables already present in  
the models. In order to do that, I simply observe wether these  
effects decrease in strength and / or lose their significance.

My question is: does any of you know a package / function in R that  
can test whether these changes in effects between models are  
significant? I figure these effects follow a T-distribution and I  
know the std. devs., so it must be easy to do manually. But I would  
like not to invent the wheel, when the function is already present.

Below is an example of what I mean. In model2, the variable z is  
added, which is hypothesized to partly explain the effect of x.  
Indeed, the effect of x decreases in model2, compared to model1. What  
I want to find out, is if this decrease is statistically significant.

Many thanks,

Rense


x <- c(1,1,1,1,1,2,2,2,2,2,3,4,4,4,5)
z <- c(2,2,2,2,2,2,2,2,3,3,3,3,4,4,5)
y <- c(1,2,2,2,3,3,3,3,4,4,4,5,5,5,5)

model1 <- lm(y~x)
model2 <- lm(y~x+z)



        [[alternative HTML version deleted]]

______________________________________________
[hidden email] 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.
Reply | Threaded
Open this post in threaded view
|

Re: [R] Compare effects between lm-models

Kuhn, Max
You can use the anova function a la:

   > anova(model1, model2)
   Analysis of Variance Table
   
   Model 1: y ~ x
   Model 2: y ~ x + z
     Res.Df    RSS Df Sum of Sq      F Pr(>F)
   1     13 4.4947                          
   2     12 4.4228  1    0.0720 0.1952 0.6665

I would suggest getting a copy of MASS and/or reading

   http://www.stats.ox.ac.uk/pub/MASS3/Exegeses.pdf


Max

   

----------------------------------------------------------------------
LEGAL NOTICE\ Unless expressly stated otherwise, this messag...{{dropped}}

______________________________________________
[hidden email] 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.
Reply | Threaded
Open this post in threaded view
|

Re: [R] Compare effects between lm-models

Rense Nieuwenhuis
Thank you Alain and Max for your swift responses.

  It might be that I'm misunderstanding your responses, but aren't  
you testing if there is a difference between the two full models?
What I want to know, os whether the effect of a specific predictor  
(x) differs between model1 and model2. I'm not interested (presently)  
if the fit of model 2 is better than that of model 1 (for instance).

thanks again,

Rense




On Jan 22, 2007, at 16:26 , Kuhn, Max wrote:

> You can use the anova function a la:
>
>> anova(model1, model2)
>    Analysis of Variance Table
>
>    Model 1: y ~ x
>    Model 2: y ~ x + z
>      Res.Df    RSS Df Sum of Sq      F Pr(>F)
>    1     13 4.4947
>    2     12 4.4228  1    0.0720 0.1952 0.6665
>
> I would suggest getting a copy of MASS and/or reading
>
>    http://www.stats.ox.ac.uk/pub/MASS3/Exegeses.pdf
>
>
> Max
>
>
>
> ----------------------------------------------------------------------
> LEGAL NOTICE
> Unless expressly stated otherwise, this message is confidential and  
> may be privileged.  It is intended for the addressee(s) only.  
> Access to this E-mail by anyone else is unauthorized.  If you are  
> not an addressee, any disclosure or copying of the contents of this  
> E-mail or any action taken (or not taken) in reliance on it is  
> unauthorized and may be unlawful.  If you are not an addressee,  
> please inform the sender immediately.
>

______________________________________________
[hidden email] 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.