CODEWITHSUNDEEP
rtimestamptime-seriesextract
Saadat
Saadat

Reputation: 51

Extracting hourly data from 15 minute data

I have a data frame with two columns, time and flow. The time interval for the time series is 15 minutes and I want to cut this time series so that the output time series has consistent one hour time intervals and the flow value from that hourly time stamp from the original data. How do I extract the hourly data?

Input:

structure(list(t = structure(c(1104555600, 1104556500, 1104557400, 
1104558300, 1104559200, 1104560100, 1104561000, 1104561900, 1104562800
), class = c("POSIXct", "POSIXt"), tzone = "EST"), flow = c(18, 
18, 18, 18.125, 18.125, 18.125, 18.125, 18.125, 18.125)), .Names = c("t", "flow"), row.names = c(NA, 9L), class = "data.frame")

And for output I would want something like

time    flow
2005-01-01 00:00:00 18.000
2005-01-01 01:00:00 18.125
2005-01-01 02:00:00 18.125

Upvotes: 1

Views: 2107

Answers (3)

NicE
NicE

Reputation: 21425

You can use cut to get the hour in which each t variable is, and then just take the first element of every cut group. If df is your dataframe:

aggregate(df, list(cut(df$t,breaks="hour")), FUN=head, 1)[,-2]

#             Group.1   flow
# 2005-01-01 00:00:00 18.000
# 2005-01-01 01:00:00 18.125
# 2005-01-01 02:00:00 18.125

Upvotes: 3

rmuc8
rmuc8

Reputation: 2989

if your dataframe is a: 

library(dplyr)
filter(a, grepl(":00:00",t))

Upvotes: 1

rmuc8
rmuc8

Reputation: 2989

You don't give any example, but from what I understand you simply want to keep every forth row.

In a data set with

time<- c(10,11,12,13,14,15,16,17,18,19)
flow<- c(3,4,5,6,7,8,9,10,11,12)
d <- data.frame(time,flow)

1    10    3
2    11    4
3    12    5
4    13    6
5    14    7
6    15    8
7    16    9
8    17   10
9    18   11
10   19   12

with

> d[seq(1, NROW(d), by = 4),]

you only keep every fourth row.

time flow
1   10    3
5   14    7
9   18   11

Upvotes: 0

Related Questions