R help - Oct 2011 - bestglm function and output in R

Hi all,

I have been trying to run a bestglm in R for a while now and am struggling
to get it to run. When I thought I had succeeded, the "output" it gave
me
was "NULL" and that's it. Below is my code:

bestglmtest<-read.table("C:\\Documents and
Settings\\clyons\\Desktop\\bestglmtest.txt",header=T,row.names=1)
> names(bestglmtest)
[1] "acc"   "age"   "sex"  
"ctmin"
> library(bestglm)
>
output<-bestglm(bestglmtest,IC="AIC",family=gaussian(link=identity))
Morgan-Tatar search since family is non-gaussian.
> bestglmtest$BestModel$coefficients
NULL

I believ my data is Gaussian so don't get the Morgan-Tatar search
"error/warning" which it threw out. Trying to find info on it was also
not
successful but I did manage to find an article with example which also had
the same message and figured out it was not an error message but a warning,
if I can say that. Asking for the best mode coefficients however, did not
provide me with any useful answers. 

What I want to do is predict how four factors/parameters (this example is
only three because I was testing the code) influence my response variable. I
have my response variable in the very last column of my text file with
predictors before it. Two of my predictors have two levels and the other two
have three levels. Perhaps this makes a difference to the ditribution type I
select? 

Any help would be greatly appreciated since I am now rightly stuck and have
no idea what else to try.

Many thanks,
Candice

--
View this message in context:
http://r.789695.n4.nabble.com/bestglm-function-and-output-in-R-tp3933547p3933547.html
Sent from the R help mailing list archive at Nabble.com.

Hi:

bestglmtest is your input data frame, is it not? From the names() line,
you can see that it has no variable named BestModel that corresponds
to a list containing a component named coefficients.

Were you perhaps looking for
output$BestModel$coefficients  ??

Dennis

On Mon, Oct 24, 2011 at 8:35 AM, Candice <candice.lyons at hotmail.com>
wrote:
> Hi all,
>
> I have been trying to run a bestglm in R for a while now and am struggling
> to get it to run. When I thought I had succeeded, the "output" it
gave me
> was "NULL" and that's it. Below is my code:
>
> bestglmtest<-read.table("C:\\Documents and
> Settings\\clyons\\Desktop\\bestglmtest.txt",header=T,row.names=1)
>> names(bestglmtest)
> [1] "acc" ? "age" ? "sex" ? "ctmin"
>> library(bestglm)
>>
output<-bestglm(bestglmtest,IC="AIC",family=gaussian(link=identity))
> Morgan-Tatar search since family is non-gaussian.
>> bestglmtest$BestModel$coefficients
> NULL
>
> I believ my data is Gaussian so don't get the Morgan-Tatar search
> "error/warning" which it threw out. Trying to find info on it was
also not
> successful but I did manage to find an article with example which also had
> the same message and figured out it was not an error message but a warning,
> if I can say that. Asking for the best mode coefficients however, did not
> provide me with any useful answers.
>
> What I want to do is predict how four factors/parameters (this example is
> only three because I was testing the code) influence my response variable.
I
> have my response variable in the very last column of my text file with
> predictors before it. Two of my predictors have two levels and the other
two
> have three levels. Perhaps this makes a difference to the ditribution type
I
> select?
>
> Any help would be greatly appreciated since I am now rightly stuck and have
> no idea what else to try.
>
> Many thanks,
> Candice
>
> --
> View this message in context:
http://r.789695.n4.nabble.com/bestglm-function-and-output-in-R-tp3933547p3933547.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.
>