Return the first position

[Inuraghe]

I have a problem, I want to create a new column that returns the first position where the number exceeds the column. How can I do it?

A B C D E Mean Pos
2 5 4 4 5 4 2
3 4 5 6 7 5 4
9 8 7 6 5 7 1

#risultati <- vector()
bjibl <- function (x){

for (i in 1:3){
for (j in 1:3600){

  if (x[i,j]>x[i, 3601]) {
    #risultati <- append(i, j, after = i)
    #risultati[i] <- j
    x1 <- as.data.frame(x[,1:j])
    x$risultati <- ncol(x1)
    i <- i +1 
    j <- 1
  }else{
    j<-j+1
  } 
} 
i <- i +1 

}
}

bjibl(db_class)

[FJCC]

Here is one method using the dplyr library.

Df <- data.frame(A=c(2,3,9),B=c(5,4,8),C=c(4,5,7),D=c(4,6,6),E=c(5,7,7),Mean=c(4,5,7))
Df
#>   A B C D E Mean
#> 1 2 5 4 4 5    4
#> 2 3 4 5 6 7    5
#> 3 9 8 7 6 7    7
library(dplyr,warn.conflicts = FALSE)
Df %>% rowwise() %>% mutate(Pos = which(c_across(A:E) > Mean)[1])
#> # A tibble: 3 x 7
#> # Rowwise: 
#>       A     B     C     D     E  Mean   Pos
#>   <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int>
#> 1     2     5     4     4     5     4     2
#> 2     3     4     5     6     7     5     4
#> 3     9     8     7     6     7     7     1

^{Created on 2020-12-09 by the reprex package (v0.3.0)}

[Inuraghe]

In my program this code doesn't work. Maybe because my dataset is huge.

I use this code but it takes too long.

db_class %>% rowwise() %>% mutate(Pos = which(c_across (1:3600) > db_class$TREDEV_STD)[1])

[FJCC]

The errors are due to the changes you made to the code. Using the data from my previous post, compare the result of these two lines.

Df %>% rowwise() %>% mutate(Pos=which(c_across(A:E) > Df$Mean)[1]) #error!
Df %>% rowwise() %>% mutate(Pos=which(c_across(A:E) > Mean)[1]) # no error

The mutate function is written so that columns are referred to by their bare names without the name of the data frame and $.

[Andrzej]

Could you please explain why do we have to use [1] in this command and what does it mean in here, local context ? I am beginner and what I know is that mutate is used for creating new column. Is this ([1]) some kind of subsetting ?

Df %>% rowwise() %>% mutate(Pos=which(c_across(A:E) > Df$Mean)[1])

[FJCC]

Yes, the square brackets are for subsetting the result returned by which(). The which() function takes a comparison and returns the indices where the comparison is true. In the example below, the comparison is x > 4 and it is true in the second and fifth positions. The which() function returns a vector with those two values. The square brackets can then be used to select from the vector that which() returns.

 x <- c(2, 5, 4, 4, 5)
 which(x > 4)
[1] 2 5

 which(x > 4)[1]
[1] 2

In the mutate() function in my earlier post, this method is used to assign to the Pos column the position of the first column that is greater than the Mean column.

[Andrzej]

Thank you very much indeed @FJCC, for detailed explanation, that helped me a lot.
best regards,
Andrzej