get_initial_values

This helper function generates a set of initial values for the numerical
optimization of the model likelihood function.

internal

Fitting (hierarchical) hidden Markov models to financial data
via maximum likelihood estimation. See Oelschläger, L. and Adam, T.
"Detecting Bearish and Bullish Markets in Financial Time Series Using
Hierarchical Hidden Markov Models" (2021, Statistical Modelling)
<doi:10.1177/1471082X211034048> for a reference on the method. A user guide
is provided by the accompanying software paper "fHMM: Hidden Markov Models
for Financial Time Series in R", Oelschläger, L., Adam, T., and Michels, R.
(2024, Journal of Statistical Software) <doi:10.18637/jss.v109.i09>.

Lennart Oelschläger

fHMM

Fitting Hidden Markov Models to Financial Data

Timo Adam

Rouven Michels

get_initial_values function

<dl><dt>data</dt>
<dd>[<code>fHMM_data</code>] 
An object of class <code>fHMM_data</code>.</dd>
<dt>ncluster</dt>
<dd>[<code>integer(1)</code>] 
Set the number of clusters for parallel optimization runs to reduce
optimization time.
By default, <code>ncluster = 1</code> (no clustering).</dd>
<dt>seed</dt>
<dd>[<code>NULL</code> | <code>integer(1)</code>] 
Set a seed for the generation of initial values.
No seed by default.</dd>
<dt>verbose</dt>
<dd>[<code>logical(1)</code>] 
Set to <code>TRUE</code> to print progress messages.</dd>
<dt>initial_estimate</dt>
<dd>[<code>NULL</code> | <code>parUncon</code>] 
Optionally defines an initial estimate for the numerical likelihood
optimization. Good initial estimates can improve the optimization process.
Can be:<ul>
<li><code>NULL</code> (the default), in this case<ul>
<li>applies a heuristic to calculate a good initial estimate</li>
<li>or uses the true parameter values (if available and
<code>data$controls$origin</code> is <code>TRUE</code>)</li>
</ul></li>
<li>or an object of class <code>parUncon</code> (i.e., a <code>numeric</code> of
unconstrained model parameters), for example the estimate of a
previously fitted model (i.e. the element <code>model$estimate</code>).</li>
</ul></dd></dl>

Arguments

Initialization of numerical likelihood optimization — get_initial_values

<dl>

<dt>data</dt>
<dd>[<code>fHMM_data</code>] 
An object of class <code>fHMM_data</code>.</dd>


<dt>ncluster</dt>
<dd>[<code>integer(1)</code>] 
Set the number of clusters for parallel optimization runs to reduce
optimization time.
By default, <code>ncluster = 1</code> (no clustering).</dd>


<dt>seed</dt>
<dd>[<code>NULL</code> | <code>integer(1)</code>] 
Set a seed for the generation of initial values.
No seed by default.</dd>


<dt>verbose</dt>
<dd>[<code>logical(1)</code>] 
Set to <code>TRUE</code> to print progress messages.</dd>


<dt>initial_estimate</dt>
<dd>[<code>NULL</code> | <code>parUncon</code>] 
Optionally defines an initial estimate for the numerical likelihood
optimization. Good initial estimates can improve the optimization process.
Can be:<ul>
<li><code>NULL</code> (the default), in this case<ul>
<li>applies a heuristic to calculate a good initial estimate</li>
<li>or uses the true parameter values (if available and
<code>data$controls$origin</code> is <code>TRUE</code>)</li>
</ul></li>
<li>or an object of class <code>parUncon</code> (i.e., a <code>numeric</code> of
unconstrained model parameters), for example the estimate of a
previously fitted model (i.e. the element <code>model$estimate</code>).</li>
</ul></dd>

</dl>

get_initial_values: Initialization of numerical likelihood optimization

Description

Usage

Value

Arguments