Returns a matrix which columns correspond to the error terms. Each
 column's row will assume value 0 if the observation was not "assigned"
 to that component.

 residuals

 internal

 sample 

Model time series using mixture autoregressive (MAR)
models. Implemented are frequentist (EM) and Bayesian
methods for estimation, prediction and model
evaluation. See Wong and Li (2002)
<doi:10.1111/1467-9868.00222>, Boshnakov (2009)
<doi:10.1016/j.spl.2009.04.009>), and the extensive
references in the documentation.

Georgi Boshnakov

mixAR

Mixture Autoregressive Models

Davide Ravagli

err function

<dl> <dt>AR</dt>
<dd>a <code>list</code> with autoregressive components.</dd> <dt>mu</dt>
<dd>Component means.</dd> <dt>y</dt>
<dd>a time series (currently a numeric vector).</dd> <dt>z</dt>
<dd>a <code>matrix</code> which columns contain observations
 "assignment". Hence each row will contain one 1 and (g-1) 0's</dd> <dt>pk</dt>
<dd>autoregressive orders.</dd></dl>

Arguments

Author

Calculate component specific error terms under MixAR model — err

<dl>

 <dt>AR</dt>
<dd>a <code>list</code> with autoregressive components.</dd>

 <dt>mu</dt>
<dd>Component means.</dd>

 <dt>y</dt>
<dd>a time series (currently a numeric vector).</dd>

 <dt>z</dt>
<dd>a <code>matrix</code> which columns contain observations
 "assignment". Hence each row will contain one 1 and (g-1) 0's</dd>

 <dt>pk</dt>
<dd>autoregressive orders.</dd>

</dl>

err: Calculate component specific error terms under MixAR model

Description

Usage

Value

Arguments

Author