[R] Marginal effects from interaction regression model

[David Winsemius]

On Jul 25, 2010, at 11:47 PM, Guillem R. wrote:

>
> As far as I know, the predict command gives the predicted values (and
> intervals) of y, but what I'm looking for is the conditional effects  
> (betas)
> of x on y conditional on values of z.

If you want the betas, then simply print the object. Don't for get the  
default parameterization is for treatment effects.

>
> I'm trying to produce a plot similar to the first shown in this link:
> http://homepages.nyu.edu/~mrg217/interaction.html#code

If you want to provide an example  ...and  a more complete description  
of the desire output, I sure someone here can finish the job of  
showing you how setting up a call to predict will get you to that goal.

-- 
David.
>
> Thanks again
>
>
>
> David Winsemius wrote:
>>
>>
>> On Jul 25, 2010, at 10:24 PM, Guillem R. wrote:
>>
>>>
>>> Dear all,
>>>
>>> I'd like to plot the marginal effect of a variable in a  
>>> multiplicative
>>> interaction regression, that is, the effect of a variable
>>> conditional on the
>>> values of another variable. As an illustration, given model lm1
>>>
>>> lm1 <- lm(y ~ x*z)
>>
>> ? predict
>>
>> Perhaps:
>>
>> predict(lm1, newdata=data.frame(x=1:10, z=5), interval="confidence")
>>
>>
>>>
>>> I'd like to get the effects of x on y conditional on the values of
>>> z, with
>>> the corresponding confidence intervals if possible. Does anyone know
>>> of any
>>> package or simple way to do this?
>>>
>>> Thanks
>>
>> -- 
>>
>> David Winsemius, MD
>> West Hartford, CT
>>
>> ______________________________________________
>> 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.
>>
>>
>
> -- 
> View this message in context: http://r.789695.n4.nabble.com/Marginal-effects-from-interaction-regression-model-tp2301858p2301884.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> 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.

David Winsemius, MD
West Hartford, CT