longitudinal.util {longitudinal}

Utility Functions for the "Longitudinal" Data Structure

Description

The above functions are all utility functions for longitudinal objects.

get.time.repeats returns the measurement design, i.e. the time points and the number of repeats per time point.

is.equally.spaced checks whether the distances between subsequent time points are all equal.

is.regularly.sampled checks whether the number of measurements are identical across time points.

has.repeated.measurements checks whether any time point as been measured more than once.

combine.longitudinal combines the measurements of two longitudinal objects. These objects must have the same (number of) variables.

condense.longitudinal condenses the multiple measurements per time point using an arbitrary function (e.g., mean, median, var).

Usage

get.time.repeats(x)
is.equally.spaced(x)
is.regularly.sampled(x)
has.repeated.measurements(x)
combine.longitudinal(x1, x2)
condense.longitudinal(x, s, func=median)

Arguments

x, x1, x2

longitudinal time series objects

s

An integer, or a vector of integers, that designate the set of time series (variables) to condense.

func

Univariate function used to summarize the multiple measurements per time point.

Value

get.time.repeats returns a list containing two vectors (time and repeats).

is.equally.spaced, is.regularly.sampled, and has.repeated.measurements return either TRUE or FALSE.

combine.longitudinal returns a longitudinal object.

condense.longitudinal returns a matrix.

Author(s)

Korbinian Strimmer (https://strimmerlab.github.io).

See Also

longitudinal, tcell.

Examples

# load "longitudinal" library
library("longitudinal")


# load tcell data set
data(tcell)
dim(tcell.34)
is.longitudinal(tcell.34)
summary(tcell.34)

# information
get.time.repeats(tcell.34)
is.equally.spaced(tcell.34)
is.regularly.sampled(tcell.34)
has.repeated.measurements(tcell.34)

# compute the mean value at each time point for the first two gene
condense.longitudinal(tcell.34, 1:2, mean)


# combine two time series
m1 <- matrix(rnorm(100), 50, 2)
m2 <- matrix(rnorm(100), 50, 2)
z1 <- as.longitudinal(m1, repeats=c(10,5,5,10,20), time=c(2,8,9,15,16))
z2 <- as.longitudinal(m2, repeats=c(10,5,5,10,20), time=c(1,8,9,15,20))

z3 <- combine.longitudinal(z1,z2)
summary(z3)
get.time.repeats(z3)  # compare with z1 and z2