MARX (version 0.2)

Simulation, Estimation, Model Selection and Forecasting for MARX Models

Description

Simulate, estimate (by t-MLE), select and forecast mixed causal-noncausal autoregressive models with possibly exogenous regressors, using methods proposed in Lanne and Saikkonen (2011) and Hecq et al. (2016) .

Copy Link

Version

Install

install.packages('MARX')

Monthly Downloads

Version

0.2

License

GPL-2

Maintainer

Sean Telg

Last Published

April 24th, 2018

Functions in MARX (0.2)

forecast.marx

Forecasting function for the MARX model

bic

The Bayesian/Schwarz information criterion (BIC) function

companion.form

Companion form function

inference

Asymptotic inference for the MARX function

commodity

Data: Monthly growth rates of commodity prices, exchange rate and industrial production index.

The Hannan-Quinn (HQ) information criterion function

ll.max

The value of the t-log-likelihood for MARX function

aic

The Akaike information criterion (AIC) function

compute.MA

Coefficients of the moving average representation function

regressor.matrix

The regressor matrix function

pseudo

The pseudo-causal model function

mixed

The MARX estimation function

selection.lag.lead

The lag-lead model selection for MARX function

selection.lag

The model selection for pseudo-ARX function

arx.ls

The ARX estimation by OLS function

marx

The MARX function

marx.t

The estimation of the MARX model by t-MLE function

sim.marx

The simulation of MARX processes