[R] How to avoid two for loops?

[conglan]

HI,

I have the following large data set:
the sample is as the following:
  Country  Product   Brand  Year_Month   DisplaySize   OperationSystem
     AE         1           20    201204              1                  1
     AE         5           20    201204              1                  1
     AE         2           28    201204              3                  2
     AE         3           27    201204              1                  1
     AE         1           20    201205              1                  1
     AE         2           28    201205              3                  2
     AE         4           20    201205              1                  2

I want to calculate the extra three columns as the following:
Let's say "20" is own brand
 #The sums of display size produced by competing brand
 #The sums of display size of other products produced by own brand
 #The sums of display size of products of competing firm in the same
operating system

I build up the following codes, apprently two loop takes too long time:

dd<-unique(cl$Year_Month)

fldata<-data.frame(NULL)

for ( i in 1:length(dd)){

  cls<-cl[cl$Year_Month==dd[i],]
  ws<-numeric(0)  #The sums of characteristics of products produced by
competing firms
  ws1<-numeric(0)  #The sums of characteristics of other products produced
by firm
  ws2<-numeric(0) #The sums of characteristics of products of competing
firm in the same operating system


  for (j in 1:dim(cls)[1]){

   wd<-sum(cls[cls$Brand!=cls$Brand[j],]$DISPLAY_SIZE_INCH)
   ws[j]<-wd
   wc<-sum(cls[cls$Brand==cls$Brand[j] & cls$Item_Id!=cls$Item_Id[j]
,]$DISPLAY_SIZE_INCH)
   ws1[j]<-wc
   wf<-sum(cls[cls$Brand!=cls$Brand[j] &
cls$OPERATING_SYST==cls$OPERATING_SYST[j]
,]$DISPLAY_SIZE_INCH)
   ws2[j]<-wf
   }

  cls<-cbind(cls,SumPCf=ws,SumOPf=ws1,SumPCfOp=ws2)
  fldata<-rbind(fldata,cls)

}


how can I avoid two loops and improve the speed of my code??????? any other
solution?


Thanks.

Tammy




--
View this message in context: http://r.789695.n4.nabble.com/How-to-avoid-two-for-loops-tp4703799.html
Sent from the R help mailing list archive at Nabble.com.
	[[alternative HTML version deleted]]