I know which.max is able to tell me the index of the first occurrence of the maximum value, however I want to obtain the maximum index value of all indexes duplicate maximum values in a given range.
Below is the data I am working with:
#Given Data Set
Dataset1 <- data.frame(Index=1:6, Value=c(.456,.92,.88,.92,.88,0.85))
Dataset1
# Index Value
1 1 0.456
2 2 0.920
3 3 0.880
4 4 0.920
5 5 0.880
6 6 0.850
#--------------------------
#Create Filtered Combinations
N=nrow(Dataset1)
Comb<-data.frame(t(combn(seq(1:(N)), 2)))
library("plyr")
library("dplyr")
FComb<-Comb %>%
filter(X2-X1>1) %>%
select(X1,X2)
FComb.mat<-as.matrix(FComb) #Filtered Combination set as matrix
colnames(FComb.mat)[1:2] <- c("ind1","ind2")
#--------------------------
#Finding maximum value from original dataset in the range between "ind1" & "ind2"
fun <- function(x,y) max(Dataset1$Value[(.x <- x:y)[-c(1, length(.x))]])
Max.Val = as.matrix(mapply(fun, FComb.mat[,1], FComb.mat[,2]))
PROBLEM:
I would like to obtain the last column "IndexMax" as shown below.
For example for ind1=1 & ind2=5 (i.e row #3), the max flow value from Dataset1 between Index=1:5 is 0.920. The index of this maximum value can be either 2 or 4, but I would like to choose 4 as it is the maximum value of the indexes.How would I implement this?
Final
# ind1 ind2 Max.Val IndexMax
1 1 3 0.92 2
2 1 4 0.92 2
3 1 5 0.92 4
4 1 6 0.92 4
5 2 4 0.88 3
6 2 5 0.92 4
7 2 6 0.92 4
8 3 5 0.92 4
9 3 6 0.92 4
10 4 6 0.88 5
I actually figured it out myself. In case this is of use to anyone else in the future:
fun2 <- function(x,y) max((which(Dataset1[,2][(.x <- x:y)][-c(1,length(.x))]==max(Dataset1[,2][(.x <- x:y)][-c(1, length(.x))])))+x)
Max.Pos=as.matrix(mapply(fun2, FComb.mat[,1],FComb.mat[,2]))
Final<-data.frame(FComb.mat,Max.Val,Max.Pos)