onlineforecast (version 1.0.0)

make_periodic: Make an forecast matrix with a periodic time signal.

Description

This function creates a data.frame with k-steps-ahead values of a periodic time signal, such that it can be added to a data.list and used inputs to a forecast model.

Usage

make_periodic(time, kseq, period, offset = 0, tstep = NA)

Arguments

time

vector of times of class "POSIXct" "POSIXt".

kseq

vector of integers, representing the desired "k-steps ahead".

period

a numeric setting the length of the period in seconds.

offset

a numeric setting an offset in the period start in seconds.

tstep

step time of k in seconds.

Value

Returns a forecast matrix (data.frame) with rownames = times, colnames = k1, k2, k3, ... The content of the data frame is the hour of day.

See Also

make_tday

Examples

# NOT RUN {
# Create a time sequence of 30 min sample period
tseq <- seq(ct("2019-01-01"), ct("2019-02-01 12:00"), by=1800)

# Make the three hourly periodic sequence
make_periodic(tseq, 1:10, 3*3600)

# With an offset of one hour
make_periodic(tseq, 1:10, 3*3600, 3600)

# }