qgarch_default_starts

Creates a compact but order-adaptive set of starting values for the selected
qgarch model.

Fits quadratic generalized autoregressive conditional
heteroskedasticity-in-mean (QGARCH-M) models motivated by Campbell and
Hentschel (1992). The package supports models with lambda fixed at zero,
lambda restricted to a function of the remaining parameters, lambda
estimated freely, and a threshold extension with state-dependent lambda.
It also provides tools for starting values, estimation, forecasting,
likelihood-ratio testing, moment diagnostics, and replication with the
included monthly U.S. stock market dataset.

Sanghyun Hong

qgarch

Quadratic GARCH-in-Mean Models for Volatility Feedback

Jedrzej Bialkowski

Moritz Wagner

qgarch_default_starts function

<dl><dt>x</dt>
<dd>A numeric vector or one-column data frame containing the observed
series.</dd>
<dt>model</dt>
<dd>Character string specifying the model variant. Must be one of
<code>"zero"</code>, <code>"restricted"</code>, <code>"free"</code>, or <code>"threshold"</code>.</dd>
<dt>arch_order</dt>
<dd>Positive integer giving the ARCH lag order.</dd>
<dt>garch_order</dt>
<dd>Positive integer giving the GARCH lag order.</dd></dl>

Arguments

Default starting values for qgarch estimation — qgarch_default_starts

<dl>

<dt>x</dt>
<dd>A numeric vector or one-column data frame containing the observed
series.</dd>


<dt>model</dt>
<dd>Character string specifying the model variant. Must be one of
<code>"zero"</code>, <code>"restricted"</code>, <code>"free"</code>, or <code>"threshold"</code>.</dd>


<dt>arch_order</dt>
<dd>Positive integer giving the ARCH lag order.</dd>


<dt>garch_order</dt>
<dd>Positive integer giving the GARCH lag order.</dd>

</dl>

qgarch_default_starts: Default starting values for qgarch estimation

Description

Usage

Value

Arguments