setup {CollocInfer}  R Documentation 
Setup Functions for proc and lik objects
Description
These functions set up lik and proc objects of squared error and multinormal processes.
Usage
LS.setup(pars,coefs=NULL,fn,basisvals=NULL,lambda,fd.obj=NULL,
more=NULL,data=NULL,weights=NULL,times=NULL,quadrature=NULL,
likfn = make.id(), likmore = NULL,eps=1e6,
posproc=FALSE,poslik=FALSE,discrete=FALSE,names=NULL,sparse=FALSE)
multinorm.setup(pars,coefs=NULL,fn,basisvals=NULL,var=c(1,0.01),fd.obj=NULL,
more=NULL,data=NULL,times=NULL,quadrature=NULL,eps=1e6,posproc=FALSE,
poslik=FALSE,discrete=FALSE,names=NULL,sparse=FALSE)
Arguments
pars 
Initial values of parameters to be estimated processes. 
coefs 
Vector giving the current estimate of the coefficients in the spline. 
fn 
A function giving the right hand side of a differential/difference equation. The function should have arguments
It should return a matrix of the same dimension of If
These functions take the same arguments as

basisvals 
Values of the collocation basis to be used. This can either be a basis object from the
For discrete systems, it may also be specified as a matrix, in which case If left as NULL, it is taken from 
lambda 
( 
var 
( 
fd.obj 
(Optional) A functional data object; if this is nonnull, 
more 
An object specifying additional arguments to 
data 
The data to be used, this can be a matrix, or a threedimensional array. If the latter, the middle dimension is taken to be replicates. The data are returned, if replicated they are returned in a concatenated form. 
weights 
( 
times 
Vector observation times for the data. If the data are replicated, times are returned in a concatenated form. 
quadrature 
Quadrature points, should contain two elements (if not NULL)

eps 
Finite differencing step size, if needed. 
posproc 
Should the state vector be constrained to be positive? If this is the case, the state is represented by
an exponentiated basis expansion in the 
poslik 
Should the state be exponentiated before being compared to the data? When the state is represented
on the log scale 
discrete 
Is this a discrete or continuoustime system? 
names 
The names of the state variables if not given by the column names of 
sparse 
Should sparse matrices be used for basis values? This option can save memory when

likfn 
Defines a map from the trajectory to the observations. This should be in the same form as

likmore 
A list containing additional inputs to 
Details
These functions provide basic setup utilities for the collocation inference methods. They define
lik
and proc
objects for sum of squared errors and multivariate normal log likelihoods with
nonlinear transfer functions describing the evolution of the state vector.
LS.setup Creates sum of squares functions
multinorm.setup Creates multinormal log likelihoods for a continuoustime system.
Value
A list with elements
coefs 
Starting values for 
lik 
The 
proc 
The 
data 
The data matrix, concatenated if from a 3d array. 
times 
The vector of observation times, concatenated if data is a 3d array. 
See Also
inneropt
, outeropt
, Profile.LS
, Profile.multinorm
, Smooth.LS
, Smooth.multinorm
Examples
# FitzHughNagumo
t = seq(0,20,0.05) # Observation times
pars = c(0.2,0.2,3) # Parameter vector
names(pars) = c('a','b','c')
knots = seq(0,20,0.2) # Create a basis
norder = 3
nbasis = length(knots) + norder  2
range = c(0,20)
bbasis = create.bspline.basis(range=range,nbasis=nbasis,norder=norder,breaks=knots)
lambda = 10000 # Penalty value
coefs = matrix(0,nbasis,2) # Coefficient matrix
profile.obj = LS.setup(pars=pars,coefs=coefs,fn=make.fhn(),basisvals=bbasis,
lambda=lambda,times=t)
# Using multinorm
var = c(1,0.01)
profile.obj = multinorm.setup(pars=pars,coefs=coefs,fn=make.fhn(),
basisvals=bbasis,var=var,times=t)
# Henon  discrete
hpars = c(1.4,0.3)
t = 1:200
coefs = matrix(0,200,2)
lambda = 10000
profile.obj = LS.setup(pars=hpars,coefs=coefs,fn=make.Henon(),basisvals=NULL,
lambda=lambda,times=t,discrete=TRUE)