[Depmix-commits] r298 - papers/jss
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Tue Jul 14 12:14:13 CEST 2009
Author: maarten
Date: 2009-07-14 12:14:07 +0200 (Tue, 14 Jul 2009)
New Revision: 298
Modified:
papers/jss/article.pdf
papers/jss/article.tex
Log:
some fixes for the jss paper
Modified: papers/jss/article.pdf
===================================================================
(Binary files differ)
Modified: papers/jss/article.tex
===================================================================
--- papers/jss/article.tex 2009-07-13 16:30:58 UTC (rev 297)
+++ papers/jss/article.tex 2009-07-14 10:14:07 UTC (rev 298)
@@ -26,20 +26,20 @@
%% an abstract and keywords
\Abstract{
- depmixS4 implements a general framework for definining and
+ \pkg{depmixS4} implements a general framework for definining and
estimating hidden Markov mixture model in the R programming language
- R2009. This includes standard Markov models,
+ \citep{R2009}. This includes standard Markov models,
latent/hidden Markov models, and latent class and finite mixture
distribution models. The models can be fitted on mixed
multivariate data with multinomial and/or gaussian distributions.
Parameters can be estimated subject to general linear constraints.
Parameter estimation is done through an EM algorithm or by a
- direct optimization approach using the Rdonlp2 optimization
+ direct optimization approach using the \pkg{Rdonlp2} optimization
routine when contraints are imposed on the parameters. A number
of illustrative examples are included.
}
-\Keywords{hidden markov model, dependent mixture model, mixture model}
+\Keywords{hidden Markov model, dependent mixture model, mixture model}
\Plainkeywords{hidden markov model, dependent mixture model, mixture model} %% without formatting
%% at least one keyword must be supplied
@@ -335,7 +335,7 @@
%Also mention use of glm, nnet and possibly other packages that we use.
-\section{Using \pkg{depmixS4}}
+\section[Using depmixS4]{Using \pkg{depmixS4}}
Two steps are involved in using \pkg{depmixS4} which are illustrated
below with examples:
@@ -364,10 +364,13 @@
the response distribution functions, and can be created with the
\code{depmix}-function as follows:
-\begin{verbatim}
-set.seed(1)
-mod <- depmix(rt~1, data=speed, nstates=2, trstart=runif(4))
-\end{verbatim}
+\begin{CodeChunk}
+\begin{CodeInput}
+> set.seed(1)
+> mod <- depmix(rt~1, data=speed, nstates=2, trstart=runif(4))
+\end{CodeInput}
+\end{CodeChunk}
+
The \code{depmix} function returns an object of class \code{depmix}
which contains the model specification (and not a fitted model!).
Note also that start values for the transition parameters are provided
@@ -375,9 +378,11 @@
provide automatic starting values.
The so-defined models needs to be \code{fit}ted with the following:
-\begin{verbatim}
-fm <- fit(mod)
-\end{verbatim}
+\begin{CodeChunk}
+\begin{CodeInput}
+> fm <- fit(mod)
+\end{CodeInput}
+\end{CodeChunk}
The \code{fit}-function returns an object of class
\code{depmix.fitted} which extends the \code{depmix} class, adding
@@ -387,27 +392,34 @@
values. These statistics may be extracted using \code{logLik},
\code{AIC} and \code{BIC}, respectively.
-\begin{verbatim}
-> fm
+\begin{CodeChunk}
+\begin{CodeInput}
+> fm
+\end{CodeInput}
+\begin{CodeOutput}
Convergence info: Log likelihood converged to within tol.
'log Lik.' -84.34175 (df=7)
AIC: 182.6835
BIC: 211.275
-\end{verbatim}
+\end{CodeOutput}
+\end{CodeChunk}
The \code{summary} method of \code{fit}ted models provides the parameter
estimates, first for the prior probabilities model, second for the
transition model, and third for the response models.
-\begin{verbatim}
+\begin{CodeChunk}
+\begin{CodeInput}
> summary(fm)
+\end{CodeInput}
+\begin{CodeOutput}
Initial state probabilties model
Model of type multinomial, formula: ~1
Coefficients:
[,1] [,2]
[1,] 0 -11.25688
Probalities at zero values of the covariates.
-0.999987 1.291797e-05
+0.999987 1.291798e-05
Transition model for state (component) 1
Model of type multinomial, formula: ~1
@@ -438,7 +450,8 @@
Coefficients:
[1] 5.511151
sd 0.1926063
-\end{verbatim}
+\end{CodeOutput}
+\end{CodeChunk}
\subsection{Transition parameters}
@@ -456,59 +469,58 @@
Covariates on the transition probabilities can be specified using a
one-sided formula as in the following example:
-\begin{verbatim}
-set.seed(1)
-mod <- depmix(rt~1, data=speed, nstates=2,
-family=gaussian(), transition=~scale(Pacc),
-instart=runif(2))
-fm <- fit(mod)
-\end{verbatim}
+\begin{CodeChunk}
+\begin{CodeInput}
+> set.seed(1)
+> mod <- depmix(rt~1, data=speed, nstates=2, family=gaussian(),
+ transition=~scale(Pacc), instart=runif(2))
+> fm <- fit(mod)
+\end{CodeInput}
+\begin{CodeOutput}
+Initial state probabilties model
+Model of type multinomial, formula: ~1
+Coefficients:
+ [,1] [,2]
+[1,] 0 10.71779
+Probalities at zero values of the covariates.
+2.214681e-05 0.9999779
-\begin{verbatim}
- Initial state probabilties model
- Model of type multinomial, formula: ~1
- Coefficients:
- [,1] [,2]
- [1,] 0 10.71779
- Probalities at zero values of the covariates.
- 2.214681e-05 0.9999779
+Transition model for state (component) 1
+Model of type multinomial, formula: ~scale(Pacc)
+Coefficients:
+ [,1] [,2]
+[1,] 0 -0.9215182
+[2,] 0 1.8649734
+Probalities at zero values of the covariates.
+0.7153513 0.2846487
- Transition model for state (component) 1
- Model of type multinomial, formula: ~scale(Pacc)
- Coefficients:
- [,1] [,2]
- [1,] 0 -0.9215182
- [2,] 0 1.8649734
- Probalities at zero values of the covariates.
- 0.7153513 0.2846487
+Transition model for state (component) 2
+Model of type multinomial, formula: ~scale(Pacc)
+Coefficients:
+ [,1] [,2]
+[1,] 0 2.471442
+[2,] 0 3.570856
+Probalities at zero values of the covariates.
+0.07788458 0.9221154
- Transition model for state (component) 2
- Model of type multinomial, formula: ~scale(Pacc)
- Coefficients:
- [,1] [,2]
- [1,] 0 2.471442
- [2,] 0 3.570856
- Probalities at zero values of the covariates.
- 0.07788458 0.9221154
+Response model(s) for state 1
- Response model(s) for state 1
+Response model for response 1
+Model of type gaussian, formula: rt ~ 1
+Coefficients:
+[1] 5.512179
+sd 0.1921098
- Response model for response 1
- Model of type gaussian, formula: rt ~ 1
- Coefficients:
- [1] 5.512179
- sd 0.1921098
+Response model(s) for state 2
- Response model(s) for state 2
+Response model for response 1
+Model of type gaussian, formula: rt ~ 1
+Coefficients:
+[1] 6.3885
+sd 0.2402693
+\end{CodeOutput}
+\end{CodeChunk}
- Response model for response 1
- Model of type gaussian, formula: rt ~ 1
- Coefficients:
- [1] 6.3885
- sd 0.2402693
-
-\end{verbatim}
-
The summary provides all parameters of the model, also the
(redundant) zeroes for the base-line category in the multinomial model.
The summary also prints the transition probabilities at the zero value
@@ -681,7 +693,7 @@
the user.
-\section{Extending \pkg{depmixS4}}
+\section[Extending depmixS4]{Extending \pkg{depmixS4}}
The \pkg{depmixS4} package was built with the aim of having the
possibility of adding new response distributions (and possibly also
@@ -767,10 +779,12 @@
object
}
)
+\end{verbatim}
+
The \code{fit} function defines a trivial \code{gamlss} model with
only an intercept to be estimated and then sets the fitted parameters
back into their respective slots in the `exgaus' object. See the help
-for \pkg{gamlss.distr} for interpretation of these parameters.
+for \code{gamlss.distr} for interpretation of these parameters.
After defining all the necessary methods for the new response model,
we can now define the dependent mixture model using this reponse.
@@ -814,7 +828,7 @@
-\section{Conclusion \& future work}
+\section[Conclusion and future work]{Conclusion \& future work}
What are our next plans?
More information about the depmix-commits
mailing list