CODEWITHSUNDEEP
rdataframe
forochelian
forochelian

Reputation: 171

How can I find the changes in a single column in R?

I have a "csv" file which includes 3 columns, 100+rows. The variables in all columns change according to the data placed in Column 1, the "Time". 

Time    Temp    Cloud
1100    22  1
1102    14  1
1104    14  2
1106    23  1
1108    12  1
1110    21  2
1112    17  2
1114    12  3
1116    24  3

I want to know when "Cloud" changes [ e.g. at 3rd and 6th row ], and I want to obtain the other variables which is placed at that row, and the row before that row. How can I do that ? Thanks

Upvotes: 0

Views: 78

Answers (2)

A_K
A_K

Reputation: 2651

I would do something like this:

df$Change <- c(0,sign(diff(df$Cloud)))
subset(df,Change!=0)[,4]

This will eliminate rows where there are no changes.

Upvotes: 1

Matthew Lundberg
Matthew Lundberg

Reputation: 42629

diff will almost do this directly. Apply it twice. Calling your example data d:

> d[c(diff(d$Cloud) != 0,FALSE) | c(FALSE, diff(d$Cloud) != 0),]
  Time Temp Cloud
2 1102   14     1
3 1104   14     2
4 1106   23     1
5 1108   12     1
6 1110   21     2
7 1112   17     2
8 1114   12     3

Upvotes: 1

Related Questions