EVOLUTION-MANAGER
Edit File: predict.arima.html
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><title>R: Forecast from ARIMA fits</title> <meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> <link rel="stylesheet" type="text/css" href="R.css" /> </head><body> <table width="100%" summary="page for predict.Arima {stats}"><tr><td>predict.Arima {stats}</td><td style="text-align: right;">R Documentation</td></tr></table> <h2>Forecast from ARIMA fits</h2> <h3>Description</h3> <p>Forecast from models fitted by <code><a href="arima.html">arima</a></code>. </p> <h3>Usage</h3> <pre> ## S3 method for class 'Arima' predict(object, n.ahead = 1, newxreg = NULL, se.fit = TRUE, ...) </pre> <h3>Arguments</h3> <table summary="R argblock"> <tr valign="top"><td><code>object</code></td> <td> <p>The result of an <code>arima</code> fit.</p> </td></tr> <tr valign="top"><td><code>n.ahead</code></td> <td> <p>The number of steps ahead for which prediction is required.</p> </td></tr> <tr valign="top"><td><code>newxreg</code></td> <td> <p>New values of <code>xreg</code> to be used for prediction. Must have at least <code>n.ahead</code> rows.</p> </td></tr> <tr valign="top"><td><code>se.fit</code></td> <td> <p>Logical: should standard errors of prediction be returned?</p> </td></tr> <tr valign="top"><td><code>...</code></td> <td> <p>arguments passed to or from other methods.</p> </td></tr> </table> <h3>Details</h3> <p>Finite-history prediction is used, via <code><a href="KalmanLike.html">KalmanForecast</a></code>. This is only statistically efficient if the MA part of the fit is invertible, so <code>predict.Arima</code> will give a warning for non-invertible MA models. </p> <p>The standard errors of prediction exclude the uncertainty in the estimation of the ARMA model and the regression coefficients. According to Harvey (1993, pp. 58–9) the effect is small. </p> <h3>Value</h3> <p>A time series of predictions, or if <code>se.fit = TRUE</code>, a list with components <code>pred</code>, the predictions, and <code>se</code>, the estimated standard errors. Both components are time series. </p> <h3>References</h3> <p>Durbin, J. and Koopman, S. J. (2001). <em>Time Series Analysis by State Space Methods</em>. Oxford University Press. </p> <p>Harvey, A. C. and McKenzie, C. R. (1982). Algorithm AS 182: An algorithm for finite sample prediction from ARIMA processes. <em>Applied Statistics</em>, <b>31</b>, 180–187. doi: <a href="https://doi.org/10.2307/2347987">10.2307/2347987</a>. </p> <p>Harvey, A. C. (1993). <em>Time Series Models</em>, 2nd Edition. Harvester Wheatsheaf. Sections 3.3 and 4.4. </p> <h3>See Also</h3> <p><code><a href="arima.html">arima</a></code> </p> <h3>Examples</h3> <pre> od <- options(digits = 5) # avoid too much spurious accuracy predict(arima(lh, order = c(3,0,0)), n.ahead = 12) (fit <- arima(USAccDeaths, order = c(0,1,1), seasonal = list(order = c(0,1,1)))) predict(fit, n.ahead = 6) options(od) </pre> <hr /><div style="text-align: center;">[Package <em>stats</em> version 3.6.0 <a href="00Index.html">Index</a>]</div> </body></html>