bsts (version 0.9.5)

aggregate.time.series: Aggregate a fine time series to a coarse summary

Description

Aggregate measurements from a fine scaled time series into a coarse time series. This is similar to functions from the xts package, but it can handle aggregation from weeks to months.

Usage

AggregateTimeSeries(fine.series,
                       contains.end,
                       membership.fraction,
                       trim.left = any(membership.fraction < 1),
                       trim.right = NULL,
                       byrow = TRUE)

Arguments

fine.series

A numeric vector or matrix giving the fine scale time series to be aggregated.

contains.end

A logical vector corresponding to fine.series indicating whether each fine time interval contains the end of a coarse time interval.

membership.fraction

A numeric vector corresponding to fine.series, giving the fraction of each time interval's observation attributable to the coarse interval containing the fine interval's first day. This will usually be a vector of 1's, unless fine.series is weekly.

trim.left

Logical indicating whether the first observation in the coarse aggregate should be removed.

trim.right

Logical indicating whether the final observation in the coarse aggregate should be removed.

byrow

Logical. If fine.series is a matrix, this argument indicates whether rows (TRUE) or columns (FALSE) correspond to time points.

Value

A matrix (if fine.series is a matrix) or vector (otherwise) containing the aggregated values of fine.series.

Examples

# NOT RUN {
  week.ending <- as.Date(c("2011-11-05",
                           "2011-11-12",
                           "2011-11-19",
                           "2011-11-26",
                           "2011-12-03",
                           "2011-12-10",
                           "2011-12-17",
                           "2011-12-24",
                           "2011-12-31"))
  membership.fraction <- GetFractionOfDaysInInitialMonth(week.ending)
  which.month <- MatchWeekToMonth(week.ending, as.Date("2011-11-01"))
  contains.end <- WeekEndsMonth(week.ending)

  weekly.values <- rnorm(length(week.ending))
  monthly.values <- AggregateTimeSeries(weekly.values, contains.end, membership.fraction)

# }