forecast.ESN

Forecast an Echo State Network (ESN) from a trained model via
 recursive forecasting. Forecast intervals are generated by simulating
 future sample path based on a moving block bootstrap of the residuals and
 estimating the quantiles from the simulations. The function is a wrapper
 for <code>forecast_esn()</code> and intended to be used in combination with
 <code>fabletools::model()</code>.

Provides a lightweight implementation of functions and methods for
fast and fully automatic time series modeling and forecasting using Echo
State Networks (ESNs).

Alexander Häußer

echos

Echo State Networks for Time Series Modeling and Forecasting

forecast.ESN function

<dl><dt>object</dt>
<dd>An object of class <code>mdl_df</code>, containing an ESN model.</dd>
<dt>new_data</dt>
<dd>Forecast horizon (n-step ahead forecast).</dd>
<dt>normal</dt>
<dd>Logical value. If <code>TRUE</code>, dist_normal() is used, otherwise dist_sample().</dd>
<dt>n_sim</dt>
<dd>Integer value. The number of future sample path generated during simulation.</dd>
<dt>specials</dt>
<dd>Currently not in use.</dd>
<dt>xreg</dt>
<dd>A <code>tsibble</code> containing exogenous variables.</dd>
<dt>...</dt>
<dd>Currently not in use.</dd></dl>

Arguments

Forecast an Echo State Network — forecast.ESN

<dl>

<dt>object</dt>
<dd>An object of class <code>mdl_df</code>, containing an ESN model.</dd>


<dt>new_data</dt>
<dd>Forecast horizon (n-step ahead forecast).</dd>


<dt>normal</dt>
<dd>Logical value. If <code>TRUE</code>, dist_normal() is used, otherwise dist_sample().</dd>


<dt>n_sim</dt>
<dd>Integer value. The number of future sample path generated during simulation.</dd>


<dt>specials</dt>
<dd>Currently not in use.</dd>


<dt>xreg</dt>
<dd>A <code>tsibble</code> containing exogenous variables.</dd>


<dt>...</dt>
<dd>Currently not in use.</dd>

</dl>

forecast.ESN: Forecast an Echo State Network

Description

Usage

Value

Arguments

Examples