EXPAR: Fitting of Exponential Autoregressive (EXPAR) Model

The amplitude-dependent exponential autoregressive (EXPAR) time series model, initially proposed by Haggan and Ozaki (1981) <doi:10.2307/2335819> has been implemented in this package. Throughout various studies, the model has been found to adequately capture the cyclical nature of datasets. Parameter estimation of such family of models has been tackled by the approach of minimizing the residual sum of squares (RSS). Model selection among various candidate orders has been implemented using various information criteria, viz., Akaike information criteria (AIC), corrected Akaike information criteria (AICc) and Bayesian information criteria (BIC). An illustration utilizing data of egg price indices has also been provided.

Getting started

Package details

AuthorSaikath Das [aut, cre], Bishal Gurung [aut], Achal Lama [aut], KN Singh [aut]
MaintainerSaikath Das <saikathdas007@gmail.com>
Package repositoryView on CRAN
Installation Install the latest version of this package by entering the following in R:

Try the EXPAR package in your browser

Any scripts or data that you put into this service are public.

EXPAR documentation built on May 29, 2024, 8:54 a.m.