auto_hpfjx: Automatic selection of the optimal HP filter with jumps and...

In jumps: Hodrick-Prescott Filter with Jumps

Automatic selection of the optimal HP filter with jumps and regressors

Description

This function needs more testing since it does not seem to work as expected. For this reasin the wrapper hpj at the moment does not allow regressors. The regularization constant for the HP filter with jumps is the maximal sum of standard deviations for the level disturbance. This value has to be passed to the hpfjx function. The auto_hpfjx runs hpfjx on a grid of regularization constants and returns the output of hpfjx selected by the chosen information criterion.

Usage

auto_hpfjx(
  y,
  X,
  grid = seq(0, sd(y) * 10, sd(y)/10),
  ic = c("bic", "hq", "aic", "aicc"),
  edf = TRUE
)

Arguments

y	numeric vector cotaining the time series;
X	numeric matrix with regressors in the columns;
grid	numeric vector containing the grid for the argument maxsum of the hpfj function;
ic	string with information criterion for the choice: the default is "bic" (simulations show this is the best choice), but also "hq", "aic" and "aicc" are available;
edf	logical scalar: TRUE (default) if the number of degrees of freedom should be computed as "effective degrees of freedom" (Efron, 1986) as opposed to a more traditional way (although not supported by theory) when FALSE.

Value

The ouput of the hpjf function corresponding to the best choice according to the selected information criterion.

Examples

y <- log(AirPassengers)
n <- length(y)
mod <- auto_hpfjx(y, trigseas(n, 12))
hpj <- ts(mod$smoothed_level, start(y), frequency = 12)
plot(y)
lines(hpj, col = "red")

jumps documentation built on April 4, 2025, 2:22 a.m.