Package 'rjd3stl'

Title: R Interface to 'JDemetra+ 3.x' time series analysis software.
Description: R Interface to 'JDemetra+ 3.x' (<https://github.com/jdemetra>) time series analysis software. It provides functions allowing to decompose a time series, including high-frequency data with multiple periodicities.
Authors: Jean Palate [aut, cre]
Maintainer: Jean Palate <[email protected]>
License: EUPL
Version: 2.1.1
Built: 2024-11-22 06:08:01 UTC
Source: https://github.com/rjdverse/rjd3stl

Help Index

Title

Description

Title

Usage

istl(
  y,
  periods,
  multiplicative = TRUE,
  swindows = NULL,
  twindows = NULL,
  ninnerloop = 1,
  nouterloop = 15,
  nojump = FALSE,
  weight.threshold = 0.001,
  weight.function = c("BIWEIGHT", "UNIFORM", "TRIANGULAR", "EPANECHNIKOV", "TRICUBE",
    "TRIWEIGHT")
)

Arguments

weight.function

Examples

q<-rjd3stl::istl(rjd3toolkit::ABS$X0.2.09.10.M, c(12, 25))
decomp<-q$decomposition

Fit a Loess regression.

Description

Fit a Loess regression.

Usage

loess(y, window, degree = 1, jump = 0)

Arguments

y

input time series.
jump

Examples

q<-rjd3stl::stlplus(rjd3toolkit::ABS$X0.2.09.10.M, 12)
decomp<-q$decomposition
t<-decomp[,'t']
matplot(cbind(t,loess(t, 121)), type='l')

Title

Description

Title

Usage

mstl(
  y,
  periods,
  multiplicative = TRUE,
  swindows = NULL,
  twindow = 0,
  ninnerloop = 1,
  nouterloop = 15,
  nojump = FALSE,
  weight.threshold = 0.001,
  weight.function = c("BIWEIGHT", "UNIFORM", "TRIANGULAR", "EPANECHNIKOV", "TRICUBE",
    "TRIWEIGHT")
)

Arguments

weight.function

Examples

q<-rjd3stl::mstl(rjd3toolkit::ABS$X0.2.09.10.M, c(12, 25))
decomp<-q$decomposition

Title

Description

Perform an STL like (based on Loess) decomposition on any periodicity

Usage

stlplus(
  y,
  period,
  multiplicative = TRUE,
  swindow = 7,
  twindow = 0,
  lwindow = 0,
  sdegree = 0,
  tdegree = 1,
  ldegree = 1,
  sjump = 0,
  tjump = 0,
  ljump = 0,
  ninnerloop = 1,
  nouterloop = 15,
  weight.threshold = 0.001,
  weight.function = c("BIWEIGHT", "UNIFORM", "TRIANGULAR", "EPANECHNIKOV", "TRICUBE",
    "TRIWEIGHT"),
  legacy = FALSE
)

Arguments

y

input time series.
period

period, any positive real number.
multiplicative

Boolean indicating if the decomposition mode is multiplicative (TRUE).
swindow

length of the seasonal filter.
twindow

length of the trend filter.
lwindow

length of the filter used to remove the trend of the seasonal
sdegree

degree of the seasonal local polynomial (0 or 1)
tdegree

degree of the trend local polynomial (0 or 1)
ldegree

degree of the low-pass local polynomial (0 or 1)
sjump

number of jumps in the computation of the seasonal
tjump

number of jumps in the computation of the trend
ljump

number of jumps in the computation of the trend in the seasonal
ninnerloop

Number of inner loops
nouterloop

Number of outer loops (computation of robust weights)
weight.threshold

Weights threshold (in [0, 0.3])
weight.function

weights function
legacy

use of the (bugged) legacy MAD

Value

A matrix with the following series: y, sa, t, s, i, fit, weights

Examples

q<-rjd3stl::stlplus(rjd3toolkit::ABS$X0.2.09.10.M, 12)
decomp<-q$decomposition