[R] Sequential multiple regression

[Bert Gunter]

See

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

?update.lm

-- Bert

On Thu, Mar 31, 2011 at 10:27 AM, Tyler Rinker <tyler_rinker at hotmail.com> wrote:
> Bert and anyone else with info,
>
> First, Bert thank you for your quick reply.  drop1 gives the results as a
> type II anova.  Is there a way to make drop1 give you type I anova (the args
> don't appear to have a way to do so)?  Another package/function perhaps?
>
> Tyler
>
>> Date: Thu, 31 Mar 2011 09:32:02 -0700
>> Subject: Re: [R] Sequential multiple regression
>> From: gunter.berton at gene.com
>> To: tyler_rinker at hotmail.com
>> CC: r-help at r-project.org
>>
>> ?drop1
>>
>> -- Bert
>>
>> On Thu, Mar 31, 2011 at 9:24 AM, Tyler Rinker <tyler_rinker at hotmail.com>
>> wrote:
>> >
>> > Hello,
>> >
>> > In the past I have tended to reside more in the ANOVA camp but am trying
>> > to become more familiar with regression techniques in R.  I would like to
>> > get the F change from a model as I take away factors:
>> >
>> > SO...
>> >
>> > mod1<-lm(y~x1+x2+x3).......mod2<-lm(y~x1,x2).......mod3<-lm(y~x1)
>> >
>> >
>> > I can do this by hand by running several models in R and taking the
>> > MSr1/MSe1, MSr2/MSe2...  This is slow and I know there's a better way.
>> >
>> > In SPSS (which I no longer use) I could easily obtain these results
>> > (F-change) as documented by Professor Andy Fields:
>> > http://www.statisticshell.com/multireg.pdf
>> >
>> > You can see the F changes for his two IV model yielding 2 F changes.
>> >  Maybe it's the language I'm using (sequential multiple regression) that
>> > yields me poor results in searching the archives and Rseek.  The results
>> > tend to be around hierarchal regression (I'm not familiar with this
>> > terminology being in the ANOVA camp).  When I look at the hier.part package
>> > and run the examples it doesn't seem to give me the F change I'm looking
>> > for.  The step function in the base program reduces the model but takes away
>> > the non sig. IV's (which is a great approach but I'm really after those F
>> > changes).  As is the usually the case I'm sure R does this simply and
>> > beautifully, I'm just not experienced with the statistical vocabulary and
>> > techniques around regression to find what I'm looking for.
>> >
>> > F-change values with R:  Any help would be appreciated.
>> >
>> > Thank you in advance,
>> > Tyler
>> >        [[alternative HTML version deleted]]
>> >
>> > ______________________________________________
>> > 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.
>> >
>>
>>
>>
>> --
>> "Men by nature long to get on to the ultimate truths, and will often
>> be impatient with elementary studies or fight shy of them. If it were
>> possible to reach the ultimate truths without the elementary studies
>> usually prefixed to them, these would not be preparatory studies but
>> superfluous diversions."
>>
>> -- Maimonides (1135-1204)
>>
>> Bert Gunter
>> Genentech Nonclinical Biostatistics
>



-- 
"Men by nature long to get on to the ultimate truths, and will often
be impatient with elementary studies or fight shy of them. If it were
possible to reach the ultimate truths without the elementary studies
usually prefixed to them, these would not be preparatory studies but
superfluous diversions."

-- Maimonides (1135-1204)

Bert Gunter
Genentech Nonclinical Biostatistics