CODEWITHSUNDEEP
rsubset
Z_D
Z_D

Reputation: 817

R - return 1st column element and column name given condition

I am having data cleaning troubles on what should be a straightforward thing and would really appreciate some help.

I have a df with dates in the first column and different categories (Red, Yellow, Orange) as column names. The remainder of the df are numbers. I would like to search through all elements of the df (other than the first column); if an element is greater than a threshold level (i >0.2, for instance), I would like to return the date and the column name. So ideally, my output would be an nx2 df like this:

  1. November, 2003 | Orange
  2. October, 1997 | Red ....

Is there a clear best method to use this: I have read about which[] and subset() and %in% and have come away without a clear answer.

Thanks again.

Upvotes: 0

Views: 75

Answers (1)

akrun
akrun

Reputation: 887901

May be

library(reshape2)
subset(melt(df, id.var='date'), value > 0.2, select=1:2)

Or using dplyr/tidyr

 library(dplyr)
 library(tidyr)
 gather(df, Var, Val,-date) %>%
                         filter(Val > 0.2) %>% 
                         select(-Val)

data

df <- structure(list(date = c("November, 2003", "October, 1997",
"September, 2005"), Red = c(0.1, 0.2, 0.3), Yellow = c(0.3, 0.4, 0.2),
 Orange = c(0.1, 0.2, 0.4)), .Names = c("date", "Red", "Yellow", "Orange"),
 row.names = c(NA, -3L), class = "data.frame")

Upvotes: 1

Related Questions