Learn R Programming

arfima

An R time series library that mixes arima models with 3 types of long-memory processes: FDWN, FGN, and PLA (power-law autocovariance.)

Copy Link

Version

Install

install.packages('arfima')

Monthly Downloads

862

Version

1.8-2

License

MIT + file LICENSE

Maintainer

JQ (Justin) Veenstra

Last Published

August 30th, 2025

Functions in arfima (1.8-2)

Plots the output from a call to tacvf

Exact log-likelihood of a long memory model

Exact log-likelihood of a long memory model with a transfer function model and series included Computes the exact log-likelihood of a long memory model with respect to a given time series as well as a transfer fucntion model and series. This function is not meant to be used directly.

The distance between modes of an arfima fit.

plot.predarfima

Plots the original time series, the predictions, and the prediction intervals for a predarfima object.

Extract Model Fitted Values

Extract Model Coefficients

Finds the best modes of an arfima fit.

The Fisher information matrix of an ARFIMA process

residuals.arfima

Extract the Residuals of a Fitted Object

Prints a tacvf object.

Extract Log-Likelihood Values

print.predarfima

Prints predictions and prediction intervals

print.summary.arfima

Prints the output of a call to summary on an arfima object

Extensive Summary of an Object

Plots the theoretical autocorralation functions (tacfs) of one or more fits.

Prints a Fitted Object

Predicts from a fitted object.

Weeds out fits from a call to arfima that are too close to each other.

sim_from_fitted

Simulate an ARFIMA time series from a fitted arfima object.

Removes a mode from an arfima fit.

Extracts the Variance-Covariance Matrix

The theoretical autocovariance function of a long memory process.

Temperature Data

Extracts the tacvfs of a fitted object

Prints changes to the package since the last update. Started in 1.4-0

Information criteria for arfima objects

Simulates, fits, and predicts persistent and anti-persistent time series. arfima

Converts AR/MA coefficients from the PACF space to operator space

Series J, Gas Furnace Data

Checks invertibility, stationarity, and identifiability of a given set of parameters

Fit ARFIMA, ARIMA-FGN, and ARIMA-PLA (multi-start) models Fits ARFIMA/ARIMA-FGN/ARIMA-PLA multi-start models to times series data. Options include fixing parameters, whether or not to fit fractional noise, what type of fractional noise (fractional Gaussian noise (FGN), fractionally differenced white noise (FDWN), or the newly introduced power-law autocovariance noise (PLA)), etc. This function can fit regressions with ARFIMA/ARIMA-FGN/ARIMA-PLA errors via the xreg argument, including dynamic regression (transfer functions).

Simulate an ARFIMA time series.

Converts AR/MA coefficients from operator space to the PACF space

Exact MLE for ARFIMA The time series is corrected for the sample mean and then exact MLE is used for the other parameters. This is a simplified version of the arfima() function that may be useful in simulations and bootstrapping.