threeRuleSmooth function

Bends a time series with a lower frequency one by smoothing their rate

threeRuleSmooth bends a time series with a time series of a lower frequency. The procedure involved is a proportional Denton benchmark.

Therefore, the resulting time series is the product of the high frequency input with a smoothed rate. This latter is extrapolated through an arithmetic sequence.

The resulting time series is equal to the low-frequency series after aggregation within the benchmark window.

threeRuleSmooth(
  hfserie,
  lfserie,
  start.benchmark = NULL,
  end.benchmark = NULL,
  start.domain = NULL,
  end.domain = NULL,
  start.delta.rate = NULL,
  end.delta.rate = NULL,
  set.delta.rate = NULL,
  ...
)

Arguments

• hfserie: the bended time series. It can be a matrix time series.
• lfserie: a time series whose frequency divides the frequency of hfserie.
• start.benchmark: an optional start for lfserie to bend hfserie. Should be a numeric of length 1 or 2, like a window for lfserie. If NULL, the start is defined by lfserie's window.
• end.benchmark: an optional end for lfserie to bend hfserie. Should be a numeric of length 1 or 2, like a window for lfserie. If NULL, the start is defined by lfserie's window.
• start.domain: an optional start of the output high-frequency series. It also defines the smoothing window : The low-frequency residuals will be extrapolated until they contain the smallest low-frequency window that is around the high-frequency domain window. Should be a numeric of length 1 or 2, like a window for hfserie. If NULL, the start is defined by hfserie's window.
• end.domain: an optional end of the output high-frequency series. It also defines the smoothing window : The low-frequency residuals will be extrapolated until they contain the smallest low-frequency window that is around the high-frequency domain window.
• start.delta.rate: an optional start for the mean of the rate difference. It is required as a common difference for the arithmetical extrapolation of the rate. Should be a numeric of length 1 or 2, like a window for lfserie. If NULL, the start is defined by lfserie's window.
• end.delta.rate: an optional end for the mean of the rate difference. It is required as a common difference for the arithmetical extrapolation of the rate. Should be a numeric of length 1 or 2, like a window for lfserie. If NULL, the end is defined by lfserie's window.
• set.delta.rate: an optional double, that allows the user to set the delta mean instead of using a mean.
• ...: if the dots contain a cl item, its value overwrites the value of the returned call. This feature allows to build wrappers.

Returns

threeRuleSmooth returns an object of class "threeRuleSmooth".

The functions plot and autoplot (the generic from ggplot2) produce graphics of the benchmarked series and the bending series. The functions in_disaggr , in_revisions , in_scatter

produce various comparisons on which plot and autoplot can also be used.

The generic accessor functions as.ts, model.list, smoothed.rate extract various useful features of the returned value.

An object of class "threeRuleSmooth" is a list containing the following components :

• benchmarked.serie: a time series, that is the result of the benchmark.

• lfrate: a time series, that is the low-frequency rate of the threeRuleSmooth.

• smoothed.rate: the smoothed rate of the threeRuleSmooth.

• hfserie.as.weights: the modified and extrapolated hfserie (see details).

• delta.rate: the low-frequency delta of the rate, used to extrapolate the low-frequenccy rate time series. It is estimated as the mean value in the specified window.

• model.list: a list containing all the arguments submitted to the function.

• call: the matched call.

Details

In order to smooth the rate, threeRuleSmooth calls bflSmooth

and uses a modified and extrapolated version of hfserie as weights :

• only the full cycles are kept
• the first and last full cycles are replicated respectively backwards and forwards to fill the domain window.

Examples

## How to use threeRuleSmooth

smooth <- threeRuleSmooth(hfserie = turnover,
                          lfserie = construction)
as.ts(smooth)
coef(smooth)
summary(smooth)
library(ggplot2)
autoplot(in_disaggr(smooth))