[R] Help using tapply with multiple variables

[Dimitris Rizopoulos]

try this:

lapply(split(dat, dat$country), function (x) {
     pres.test(x$pop1995, x$pop2000)
})


I hope it helps.

Best,
Dimitris

Corey Sparks wrote:
> Dear list,
> I have the function (as a simple example, which is actually part of a 
> larger function)
> 
> pres.test<-function(N0=N0, N1=N1)
> {
>     dt<-5
>     r<-log(N1/N0)/dt
>     r
> }
> 
> which calculates the annual growth rates in a population
> 
> Where N0 is the population classified into age intervals, say 5 years, 
> at time=1995, and N1 is the population by 5 year age classes at time=2000.
> 
> For example some data like this are:
> country   pop1995 pop2000
> 1     1712    1940
> 1     1329    1535
> 1     1101    1268
> 1     911    1048
> 1      758     859
> 2      627     710
> 2      513     584
> 2      420     475
> 2     754     965
> 2     638     741
> 
> I want to use the tapply function to apply the pres.test function over 
> all countries in my data table (the real data have ~150 countries and 10 
> age classes), but I can't seem to get tapply to take as input to FUN 
> more than one variable, for example I tried:
> 
> popdat<-cbind(dat2.sub$pop1995, dat2.sub$pop2000) #try to force the two 
> time points into a vector form
> 
> tapply(popdat, dat2.sub$country, pres.test, ...=list(N0=popdat[,1], 
> N1=popdat[,2])) #tried to pass the other arguments that pres.test needs  
> via ...
> 
> And got the error:
> Error in tapply(popdat, dat2.sub$Country, pres.test, ... = list(N0 = 
> popdat[,  :
>   arguments must have same length
> 
> I see that the function mapply will take multiple arguements, but I 
> don't think it will use an INDEX like tapply.
> 
> Any comments or clarification would be most appreciated.
> 
> Corey
> 

-- 
Dimitris Rizopoulos
Assistant Professor
Department of Biostatistics
Erasmus Medical Center

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