group_times {spatsoc}

Group Times

Description

group_times groups rows into time groups. The function accepts date time formatted data and a threshold argument. The threshold argument is used to specify a time window within which rows are grouped.

Usage

group_times(DT = NULL, datetime = NULL, threshold = NULL)

Arguments

DT

input data.table

datetime

name of date time column(s). either 1 POSIXct or 2 IDate and ITime. e.g.: 'datetime' or c('idate', 'itime')

threshold

threshold for grouping times. e.g.: '2 hours', '10 minutes', etc. if not provided, times will be matched exactly. Note that provided threshold must be in the expected format: '## unit'

Details

The DT must be a data.table. If your data is a data.frame, you can convert it by reference using data.table::setDT.

The datetime argument expects the name of a column in DT which is of type POSIXct or the name of two columns in DT which are of type IDate and ITime.

threshold must be provided in units of minutes, hours or days. The character string should start with an integer followed by a unit, separated by a space. It is interpreted in terms of 24 hours which poses the following limitations:

• minutes, hours and days cannot be fractional

• minutes must divide evenly into 60

• minutes must not exceed 60

• minutes, hours which are nearer to the next day, are grouped as such

• hours must divide evenly into 24

• multi-day blocks should divide into the range of days, else the blocks may not be the same length

In addition, the threshold is considered a fixed window throughout the time series and the rows are grouped to the nearest interval.

If threshold is NULL, rows are grouped using the datetime column directly.

Value

group_times returns the input DT appended with a timegroup column and additional temporal grouping columns to help investigate, troubleshoot and interpret the timegroup.

The actual value of timegroup is arbitrary and represents the identity of a given timegroup which 1 or more individuals are assigned to. If the data was reordered, the group may change, but the contents of each group would not.

The temporal grouping columns added depend on the threshold provided:

• threshold with unit minutes: "minutes" column added identifying the nearest minute group for each row.

• threshold with unit hours: "hours" column added identifying the nearest hour group for each row.

• threshold with unit days: "block" columns added identifying the multiday block for each row.

A message is returned when any of these columns already exist in the input DT, because they will be overwritten.

See Also

group_pts group_lines group_polys

Examples

# Load data.table
library(data.table)


# Read example data
DT <- fread(system.file("extdata", "DT.csv", package = "spatsoc"))

# Cast the character column to POSIXct
DT[, datetime := as.POSIXct(datetime, tz = 'UTC')]

group_times(DT, datetime = 'datetime', threshold = '5 minutes')

group_times(DT, datetime = 'datetime', threshold = '2 hours')

group_times(DT, datetime = 'datetime', threshold = '10 days')