FunctionalModel-class: A FunctionalModel for a Functional Model

In thomasWeise/regressoR.functional.models: An R Package Providing an Blueprints and Tools for Functional Models for Regression

Description

This class holds the a blueprint for a functional model, i.e., an unparameterized function model. Such a model is defined by a function f which accepts one scalar input x and a parameterization vector par and returns an output scalar y. The model depends on the parameterization par, which will later be subject to optimization to make the function f(x, par) fit to a model dataset (x, y).

Slots

f

the model function, taking as parameters a value x followed by a parameter vector par

estimator

is a function which takes in a vector of x and a vector of y values and returns an estimate of the parameters, or NULL if no estimate can be made better than just standard random numbers

gradient

a function which takes in a value x and par and returns a vector with the gradient for each parameter dimension

paramCount

the number of model parameters

paramLower

the lower bounds for the parameters, or NULL if none are required. An element of the vector may be set of NA if no lower limit for that limit is specified (while lower limits are given for other parameter values).

paramUpper

the upper bounds for the parameters, or NULL if none are required. An element of the vector may be set of NA if no lower upper for that limit is specified (while upper limits are given for other parameter values).

name

a textual name of the model

See Also

FunctionalModel.new

thomasWeise/regressoR.functional.models documentation built on May 17, 2019, 8:45 p.m.