PointProcessModel-class: Class "PointProcessModel"

In ppstat: Point Process Statistics

Description

An object of class "PointProcessModel" is returned from the constructor pointProcessModel. It contains the fitted generalized linear point process model. The class inherits class "PointProcess".

Objects from the Class

Objects can be created by calls of the form pointProcessModel(formula, data, family) where data is an object of class MarkedPointProcess.

The model object contains the data used for fitting the model and a formula specification of the model. When the object is the result of calling pointProcessModel the object contains the fitted model parameters (via MLE) and the estimated covariance matrix for the parameters.

A model matrix is by default computed from the data based on the formula when the object is created using pointProcessModel. Each row in the matrix corresponds to one grid point in the data set.

Slots

basisEnv:

an "environment" containing a list with basis function evaluations.

basisPoints:

a "numeric" containing the evaluation points for the basis functions.

coefficients:

a "numeric" vector containing the estimated model parameters.

crossProd:

a "list" containing the cross product ot the model matrix.

filterTerms:

a "numeric" specifying the terms in the formula that are filter terms.

modelMatrixCol:

a numeric vector.

modelMatrixEnv:

an environment that contains an object of class "Matrix" - the model matrix.

lambda

a "numeric". The vector of penalization weights. A length 0 vector means no penalization.

optimResult:

a "list" containing the results from numerical optimization (optim).

responseMatrix:

a "Matrix". The model matrix subsetted to the point observations for the response.

var:

a "matrix". The estimated variance matrix of the model parameter estimator.

varMethod:

a "character" specifying the method used for estimating the variance matrix.

Methods

coefficients

signature(object = "PointProcessModel"): Returns the estimated model parameters.

coefficients<-

signature(.Object = "PointProcessModel",value = "numeric"): Sets the model parameters.

computeBasis

signature(model = "PointProcessModel", term = "character", varLabels = "character"): Computes the basis function evaluations.

computeDDMinusLogLikelihood

signature(model = "PointProcessModel"): Computes the hessian of the minus-log-likelihood function.

computeDMinusLogLikelihood

signature(model = "PointProcessModel"): Computes the gradient of the minus-log-likelihood function.

computeLinearPredictor

signature(model = "PointProcessModel"): Computes the linear predictor.

computeModelMatrix

signature(model = "PointProcessModel"): Computes the model matrix.

getModelMatrix

signature(model = "PointProcessModel"): Returns the model matrix.

getLinearFilter

signature(model = "PointProcessModel")(model, se = FALSE, nr, ...): Returns the values of the linear filter functions as a data.frame. If se = TRUE, the functions returns a list with the linear filter functions in the first component and the estimated standard errors in the second component.

setModelMatrix<-

signature(model = "PointProcessModel",value = "list"): Sets the model matrix.

penalty

signature(model = "PointProcessModel",): Returns the lambda vector of penalty weights.

penalty<-

signature(model = "PointProcessModel", "numeric"): Sets the lambda vector of penalty weights.

predict

signature(object = "PointProcessModel"): Returns the intensity. This is the linear predictor, as computed by computeLinearPredictor, transformed by phi.

termPlot

signature(model = "PointProcessModel"): Plots the (estimated) linear filter functions.

show

signature(object = "PointProcessModel"): Prints the model object.

summary

signature(object = "PointProcessModel"): Summarizes the model object. See summary.

subset

signature(x = "PointProcessModel"): Creates a new point process model based on a subset of the data by passing the argument(s) to the subset method for the process data, see MarkedPointProcess. The family and formula are retained in the new model.

update

signature(model = "PointProcessModel"): See update.

vcov

signature(model = "PointProcessModel"): Returns the estimated variance matrix of the parameter estimator.

Author(s)

Niels Richard Hansen, [e-mail]

See Also

pointProcessModel, PointProcess, PointProcessSmooth, PointProcessKernel ProcessData, summary, formula.

Examples

showClass("PointProcessModel")

ppstat documentation built on May 2, 2019, 5:26 p.m.