mstep.norm

Re-estimates the parameters of a normal emission distribution as part of the EM algorithm for HMMs and HSMMs.
 This is called by the <code>hmm</code> and <code>hsmm</code> functions. It is a suitable prototype function for users
 wishing to design their own emission distributions.

Parameter estimation and prediction for hidden Markov and semi-Markov models for data with multiple observation sequences.  Suitable for equidistant time series data, with multivariate and/or missing data. Allows user defined emission distributions.

Jared OConnell

mhsmm

Inference for Hidden Markov and Semi-Markov Models

mstep.norm function

<dl> <dt>x</dt>
<dd>A vector of observed values</dd> <dt>wt</dt>
<dd>A T x J matrix of weights. Column entries are the weights for respective states.</dd></dl>

Arguments

Jared O'Connell jaredoconnell@gmail.com

Author

Performs re-estimation (the M-step) for a normal emission distribution — mstep.norm

<dl>

 <dt>x</dt>
<dd>A vector of observed values</dd>

 <dt>wt</dt>
<dd>A T x J matrix of weights. Column entries are the weights for respective states.</dd>

</dl>

mstep.norm: Performs re-estimation (the M-step) for a normal emission distribution

Description

Usage

Value

Arguments

Author

Details