SuanShu, a Java numerical and statistical library

Package com.numericalmethod.suanshu.stats.timeseries.linear.univariate.arima

Class Summary
ARIMAForecast Forecasts an ARIMA time series using the innovative algorithm.
ARIMAForecast.Forecast The forecast value and variance.
ARIMAForecastMultiStep Makes forecasts for a time series assuming an ARIMA model using the innovative algorithm.
ARIMAModel An ARIMA(p, d, q) process, Xt, is such that \[ (1 - B)^d X_t = Y_t \] where B is the backward or lag operator, d the order of difference, Yt an ARMA(p, q) process, for which \[ Y_t = \mu + \Sigma \phi_i Y_{t-i} + \Sigma \theta_j \epsilon_{t-j} + \epsilon_t, \]
ARIMASim This class simulates an ARIMA (AutoRegressive Integrated Moving Average) process.
ARIMAXModel The ARIMAX model (ARIMA model with eXogenous inputs) is a generalization of the ARIMA model by incorporating exogenous variables.
AutoARIMAFit Selects the order and estimates the coefficients of an ARIMA model automatically by AIC or AICC.

Copyright © 2010-2015 Numerical Method Incorporation Limited. All Rights Reserved.