[R] Re :argument is not numeric or logical

[David Winsemius]

On May 2, 2010, at 4:25 PM, David Winsemius wrote:

>
> On May 2, 2010, at 3:27 PM, Mohan L wrote:
>
>>
>>
>> >as.numeric(assame$Login)  //convert to numerics
>>
>>
>> NOOOOooooooo. You did not do what what I suggested earlier. Do not  
>> reach for your keyboard in the R console until you have read the  
>> FAQ section regarding converting factors to numeric. If you don't  
>> want to read the FAQ
>>
>> Hi David,
>> Now I understand the trick :
>>
>> as.numeric(as.character(Login))
>>
>>
>> > avglog <-  
>> with(assame,tapply(as.numeric(as.character(Login)),stdate,mean))> >  
>> avglog
>> 01/11/09 00:00 02/11/09 00:00 04/11/09 00:00 05/11/09 00:00  
>> 06/11/09 00:00
>>      22.04866       16.65358       15.84970       16.58600        
>> 16.20743
>> 07/11/09 00:00
>>      18.52379
>>
>> But still it will not work :
>>
>> > avglog <- with(assame,tapply(as.numeric(levels(Login) 
>> [as.integer(Login)]),stdate,mean))

>
> Sometimes using with(. , .) does not work when the variable name is  
> buried deep in a nested set of functions. Furthermore mean needs an  
> na.rm argument if, as I suspect, you now have some some NA's in the  
> new variable. What happens when you:
>
> assame$Login.num <- as.numeric(as.character(assame$Login))
> avglog <-
> avglog
>

Meant to type:

avglog <- with(assame,  tapply(Login.num, stdate, mean, na.rm=TRUE) )

>
> Untested of course since we STILL have no reproducible example. Had  
> you used dput on head(sample, 40) or head(sample2, 40) or  
> head(assame, 40) we would not still be here guessing at what you  
> really are working with.
>
> -- 
> David.
>
>> Error in tapply(as.numeric(levels(Login)[as.integer(Login)]),  
>> stdate,  :
>>  arguments must have same length
>>
>>
>> Thanks & Rg
>> Mohan L
>>
>>
>>
>>
>
> 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.

David Winsemius, MD
West Hartford, CT