fArma-package | R Documentation |
The Rmetrics "fArma" package is a collection of functions to analyze and model ARMA time series processes which special emphesis in Finance.
Package: \tab fArma\cr Type: \tab Package\cr Version: \tab R 3.0.1\cr Date: \tab 2014\cr License: \tab GPL Version 2 or later\cr Copyright: \tab (c) 1999-2014 Rmetrics Assiciation\cr URL: \tab \url{https://www.rmetrics.org}
The 'fARMA' package is a collection of functions to analyze, to simulate, to fit parameteres, and to forecast ARMA model and long range dependency of fincnaial time series models.
The section provides a collection simple to use functions to model univariate autoregressive moving average time series processes, including time series simulation, parameter estimation, diagnostic analysis of the fit, and predictions of future values.
armaSim simulates an artificial ARMA time series process armaFit fits the parameters of an ARMA time series process
Extractor Functions:
fitted returns fitted values coef returns coefficients residuals returns residuals
Forecasting Function:
predict forecasts and optionally plots ARMA processes
Generic print, plot and summary functions:
print print method plot plot method summary summary method
Here we provide two functions to compute the statistics of a true ARMA time series process.
armaRoots roots of the characteristic ARMA polynomial armaTrueacf true autocorrelation function of an ARMA process
This is a collection and description of functions to investigate the long range dependence or long memory behavior of an univariate time series process. Included are functions to simulate fractional Gaussian noise and fractional ARMA processes, and functions to estimate the Hurst exponent by several different methods.
Functions to simulate long memory time series processes:
fnmSim simulates fractional Brownian motion - mvn from the numerical approximation of the stochastic integral - chol from the Choleski's decomposition of the covariance matrix - lev using the method of Levinson - circ using the method of Wood and Chan - wave using the wavelet synthesis
fgnSim simulates fractional Gaussian noise - beran using the method of Beran - durbin using the method Durbin and Levinson - paxson using the method of Paxson
farimaSim simulates FARIMA time series processes
Functions to estimate the Hurst exponent:
aggvarFit aggregated variance method diffvarFit differenced aggregated variance method absvalFit aggregated absolute value (moment) method higuchiFit Higuchi's or fractal dimension method pengFit Peng's or variance of residuals method rsFit R/S Rescaled Range Statistic method perFit periodogram method boxperFit boxed (modified) periodogram method whittleFit Whittle estimator hurstSlider interactive Display of Hurst Estimates
Function for the wavelet estimator:
waveletFit wavelet estimator
This section provides two sets of functions functions to investigate
the true statistics of the long range dependence or long memory behavior
of univariate FGN or FARIMA time series processes.
FGN Models:
fgnTrueacf returns true FGN covariances fgnTruefft returns true FGN fast Fourier transform fgnStatsSlider returns a plot of true FGN Statistics
FARIMA Models:
farimaTrueacf returns true FARIMA covariances farimaTruefft returns true FARIMA fast Fourier transform farimaStatsSlider returns a plot of true FARIMA Statistics
The fArma
Rmetrics package is written for educational
support in teaching "Computational Finance and Financial Engineering"
and licensed under the GPL.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.