SCM

SCM Repository

[depmix] Annotation of /trunk/man/response.Rd
ViewVC logotype

Annotation of /trunk/man/response.Rd

Parent Directory Parent Directory | Revision Log Revision Log


Revision 86 - (view) (download) (as text)

1 : ingmarviss 39 \name{response}
2 : ingmarviss 82
3 : ingmarviss 39 \docType{method}
4 : ingmarviss 82
5 : ingmarviss 39 \alias{response}
6 : ingmarviss 78 \alias{transInit}
7 :     \alias{GLMresponse}
8 : ingmarviss 82
9 :     \alias{transInit,formula,GLMresponse-method}
10 :     \alias{GLMresponse,formula,GLMresponse-method}
11 :    
12 : ingmarviss 39 \title{Methods for creating depmix response models}
13 : ingmarviss 82
14 : ingmarviss 39 \description{Create \code{\link[depmix]{response}} models.}
15 : ingmarviss 82
16 : ingmarviss 39 \usage{
17 :    
18 : ingmarviss 86 GLMresponse(formula, data=NULL, family=gaussian(), pstart=NULL,
19 :     fixed=NULL, prob=TRUE, ...)
20 : ingmarviss 39
21 : ingmarviss 82 transInit(formula, nstates, data=NULL, family=multinomial(),
22 : ingmarviss 86 pstart=NULL, fixed=NULL, prob=TRUE, ...)
23 : ingmarviss 82
24 : ingmarviss 39 }
25 : ingmarviss 82
26 : ingmarviss 39 \arguments{
27 :     \item{formula}{A model formula.}
28 : ingmarviss 82 \item{nstates}{The number of states of the model.}
29 : ingmarviss 39 \item{family}{A family object; currently only multinomial() and
30 :     gaussian() are allowed options.}
31 :     \item{data}{An optional data.frame to interpret the variables from the
32 :     formula argument in.}
33 :     \item{pstart}{Starting values for the coefficients and other
34 :     parameters, ie the standard deviation for the gaussian() family.}
35 :     \item{fixed}{Logical vector indicating which paramters are to be fixed.}
36 :     \item{prob}{Logical indicating whether the starting values for
37 :     multinomial() family models are probabilities or logistic
38 :     parameters (see details).}
39 : ingmarviss 86 \item{...}{Not used currently.}
40 : ingmarviss 39 }
41 : ingmarviss 82
42 : ingmarviss 39 \details{
43 :    
44 :     Both methods use the familiar formula interface from linear and general
45 :     linear models to specify how responses (or transition or prior
46 :     parameters) depend on covariates.
47 :    
48 :     The GLMresponse model is an interface to the glm functions of which
49 :     the functionality is leant, ie predict, fit and density functions.
50 :    
51 :     The transInit response model provides functionality for multinomial
52 :     responses that are currently fit using nnet (this may change in the
53 :     future but this should not affect the interface of this function). Note
54 :     that the transInit model actually lacks a reponse, ie the y-slot is
55 :     empty, at the time of construction, as the transitions are not
56 :     observed.
57 :    
58 :     }
59 : ingmarviss 82
60 : ingmarviss 39 \value{
61 :     \code{GLMresponse} and \code{transInit} return objects of class
62 :     \code{GLMresponse} and \code{transInit} respectively; both classes
63 :     extend the \code{\link[depmix]{response-class}}.
64 :    
65 : ingmarviss 82 For extensibility, objects with class \code{response} have at least
66 :     methods: \code{predict} to return \code{predict}'ed responses,
67 :     \code{dens} to return the \code{dens}'ity of responses.
68 : ingmarviss 39 }
69 : ingmarviss 82
70 :     \author{Ingmar Visser}
71 :    
72 :     \keyword{methods}

root@r-forge.r-project.org
ViewVC Help
Powered by ViewVC 1.0.0  
Thanks to:
Vienna University of Economics and Business Powered By FusionForge