R help - Feb 2010 - creating functions question

Hi All,

I am interested in creating a function that will take x number of  lm
objects and automate the comparison of each model (using anova). Here
is a simple example (the actual function will involve more than what
Im presenting but is irrelevant for the example):

# sample data:

id<-rep(1:20)
n<-c(10,20,13,22,28,12,12,36,19,12,36,75,33,121,37,14,40,16,14,20)
r<-c(.68,.56,.23,.64,.49,-.04,.49,.33,.58,.18,-.11,.27,.26,.40,.49,
.51,.40,.34,.42,.16)
mod2<-factor(c(rep(c(1,2,3,4),5)))
da<-data.frame(id, n, r, mod1, mod2)


reg0<-lm(da$r ~ 1)
reg1<-lm(da$r ~ da$mod1)
reg2<-lm(da$r ~ da$mod1 + da$mod2)

# This is as far as I get with the function:

MRfit <- function( ...) {
 models <- list(...)
 fit<- anova(models)
 return(fit)
}

MRfit(reg0,reg1,reg2)

# This is what I get from R:

#  Error in UseMethod("anova") :
#  no applicable method for 'anova' applied to an object of class
"list"

# which makes sense, but Im not sure how to work with this list
considering the desired output.
# Ideally I would like to get:

MRfit(reg0,reg1,reg2)   # I know that I could just do anova(reg0,reg1,
reg2) but the actual function will have additional functions
integrated

Analysis of Variance Table

Model 1: da$r ~ 1
Model 2: da$r ~ da$mod1
Model 3: da$r ~ da$mod1 + da$mod2

 Res.Df     RSS Df Sum of Sq      F Pr(>F)
1     19 0.83848
2     18 0.78668  1  0.051803 1.1230 0.3060
3     15 0.69193  3  0.094747 0.6847 0.5752


Any help is much appreciated!

Thank you,

AC

try this:

MRfit <- function (...) {
     models <- list(...)
     do.call(anova, models)
}


I hope it helps.

Best,
Dimitris


AC Del Re wrote:
> Hi All,
> 
> I am interested in creating a function that will take x number of  lm
> objects and automate the comparison of each model (using anova). Here
> is a simple example (the actual function will involve more than what
> Im presenting but is irrelevant for the example):
> 
> # sample data:
> 
> id<-rep(1:20)
> n<-c(10,20,13,22,28,12,12,36,19,12,36,75,33,121,37,14,40,16,14,20)
> r<-c(.68,.56,.23,.64,.49,-.04,.49,.33,.58,.18,-.11,.27,.26,.40,.49,
> .51,.40,.34,.42,.16)
> mod2<-factor(c(rep(c(1,2,3,4),5)))
> da<-data.frame(id, n, r, mod1, mod2)
> 
> 
> reg0<-lm(da$r ~ 1)
> reg1<-lm(da$r ~ da$mod1)
> reg2<-lm(da$r ~ da$mod1 + da$mod2)
> 
> # This is as far as I get with the function:
> 
> MRfit <- function( ...) {
>  models <- list(...)
>  fit<- anova(models)
>  return(fit)
> }
> 
> MRfit(reg0,reg1,reg2)
> 
> # This is what I get from R:
> 
> #  Error in UseMethod("anova") :
> #  no applicable method for 'anova' applied to an object of class
"list"
> 
> # which makes sense, but Im not sure how to work with this list
> considering the desired output.
> # Ideally I would like to get:
> 
> MRfit(reg0,reg1,reg2)   # I know that I could just do anova(reg0,reg1,
> reg2) but the actual function will have additional functions
> integrated
> 
> Analysis of Variance Table
> 
> Model 1: da$r ~ 1
> Model 2: da$r ~ da$mod1
> Model 3: da$r ~ da$mod1 + da$mod2
> 
>  Res.Df     RSS Df Sum of Sq      F Pr(>F)
> 1     19 0.83848
> 2     18 0.78668  1  0.051803 1.1230 0.3060
> 3     15 0.69193  3  0.094747 0.6847 0.5752
> 
> 
> Any help is much appreciated!
> 
> Thank you,
> 
> AC
> 
> ______________________________________________
> 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.
> 
-- 
Dimitris Rizopoulos
Assistant Professor
Department of Biostatistics
Erasmus University Medical Center

Address: PO Box 2040, 3000 CA Rotterdam, the Netherlands
Tel: +31/(0)10/7043478
Fax: +31/(0)10/7043014

Perfect!

 Thank you, Dimitris.

 AC


 On Mon, Feb 15, 2010 at 1:45 PM, Dimitris Rizopoulos
 <d.rizopoulos at erasmusmc.nl> wrote:
> try this:
>
> MRfit <- function (...) {
> ? ?models <- list(...)
>> ? ?do.call(anova, models)
>> }
>>
>>
>> I hope it helps.
>>
>> Best,
>> Dimitris
>>
>>
>> AC Del Re wrote:
>>>
>>> Hi All,
>>>
>>> I am interested in creating a function that will take x number of
?lm
>>> objects and automate the comparison of each model (using anova).
Here
>>> is a simple example (the actual function will involve more than
what
>>> Im presenting but is irrelevant for the example):
>>>
>>> # sample data:
>>>
>>> id<-rep(1:20)
>>>
n<-c(10,20,13,22,28,12,12,36,19,12,36,75,33,121,37,14,40,16,14,20)
>>>
r<-c(.68,.56,.23,.64,.49,-.04,.49,.33,.58,.18,-.11,.27,.26,.40,.49,
>>> .51,.40,.34,.42,.16)
>>> mod2<-factor(c(rep(c(1,2,3,4),5)))
>>> da<-data.frame(id, n, r, mod1, mod2)
>>>
>>>
>>> reg0<-lm(da$r ~ 1)
>>> reg1<-lm(da$r ~ da$mod1)
>>> reg2<-lm(da$r ~ da$mod1 + da$mod2)
>>>
>>> # This is as far as I get with the function:
>>>
>>> MRfit <- function( ...) {
>>> ?models <- list(...)
>>> ?fit<- anova(models)
>>> ?return(fit)
>>> }
>>>
>>> MRfit(reg0,reg1,reg2)
>>>
>>> # This is what I get from R:
>>>
>>> # ?Error in UseMethod("anova") :
>>> # ?no applicable method for 'anova' applied to an object of
class "list"
>>>
>>> # which makes sense, but Im not sure how to work with this list
>>> considering the desired output.
>>> # Ideally I would like to get:
>>>
>>> MRfit(reg0,reg1,reg2) ? # I know that I could just do
anova(reg0,reg1,
>>> reg2) but the actual function will have additional functions
>>> integrated
>>>
>>> Analysis of Variance Table
>>>
>>> Model 1: da$r ~ 1
>>> Model 2: da$r ~ da$mod1
>>> Model 3: da$r ~ da$mod1 + da$mod2
>>>
>>> ?Res.Df ? ? RSS Df Sum of Sq ? ? ?F Pr(>F)
>>> 1 ? ? 19 0.83848
>>> 2 ? ? 18 0.78668 ?1 ?0.051803 1.1230 0.3060
>>> 3 ? ? 15 0.69193 ?3 ?0.094747 0.6847 0.5752
>>>
>>>
>>> Any help is much appreciated!
>>>
>>> Thank you,
>>>
>>> AC
>>>
>>> ______________________________________________
>>> 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.
>>>
>>
>> --
>> Dimitris Rizopoulos
>> Assistant Professor
>> Department of Biostatistics
>> Erasmus University Medical Center
>>
>> Address: PO Box 2040, 3000 CA Rotterdam, the Netherlands
>> Tel: +31/(0)10/7043478
>> Fax: +31/(0)10/7043014
>>
>