select_random_subset: Select Random

In majpark21/TSexploreR: Explorative Tools for Time Series

Description

Select a random subset of time series from a dataset by using unique ID subset. Optionally, a minimum number of observation for the individual IDs can be provided.

Usage

select_random_subset(in.file, ntraj, col.uniqID, col.time,
  min.obs = NULL, col.whatokeep = NULL, out.file = NULL)

Arguments

in.file	A character or a variable for the current environment. If a character, a path to a .csv file is expected.
ntraj	Numeric. How many unique IDs to extract?
col.uniqID	Character. Name of the column with unique IDs.
col.time	Character. Name of the column with time.
min.obs	Optional numeric. Minimum number of observations of an ID.
col.whatokeep	Optional character. Name of the columns to be returned in the output. If NULL returns all columns. Columns with uniqID and time are always returned.
out.file	Optional character. If provided the output is saved as a .csv file to the path indicated by out.file. Otherwise, the new dataset is returned.

Value

A data.table with the columns indicated in col.whatokeep.

Examples

# Simulate 10 phase-shifted sinusoids with 3 different level of noises
# ("experimental conditions", first grouping).
x <- multi_sims(type = "ps", noises = c(0.2,0.4), n = 10)
plot_sim(x)

# Read/export 5 random TS from/in environment
x[, uniqID := paste(noise, variable, sep ="_")]
x_subset <- select_random_subset(in.file=x, ntraj=5, col.uniqID="uniqID", col.time="Time")

majpark21/TSexploreR documentation built on Oct. 16, 2019, 2:46 p.m.