srtree-2.0.0.2: A general library to work with Symbolic Regression expression trees.
Copyright(c) Fabricio Olivetti 2021 - 2024
LicenseBSD3
Maintainer[email protected]
Stabilityexperimental
PortabilityConstraintKinds
Safe HaskellNone
LanguageHaskell2010

Algorithm.SRTree.ConfidenceIntervals

Description

Functions to optimize the parameters of an expression.

Synopsis

Documentation

data PType Source #

profile likelihood algorithms: Bates (classical), ODE (faster), Constrained (fastest) The Constrained approach returns only the endpoints.

Constructors

Bates 
ODE 
Constrained 

Instances

Instances details
Read PType Source # 
Instance details

Defined in Algorithm.SRTree.ConfidenceIntervals

Show PType Source # 
Instance details

Defined in Algorithm.SRTree.ConfidenceIntervals

Methods

showsPrec :: Int -> PType -> ShowS #

show :: PType -> String #

showList :: [PType] -> ShowS #

data CIType Source #

Confidence Interval using Laplace approximation or profile likelihood.

data BasicStats Source #

Basic stats of the data: covariance of parameters, correlation, standard errors

Constructors

MkStats 

Instances

Instances details
Show BasicStats Source # 
Instance details

Defined in Algorithm.SRTree.ConfidenceIntervals

Eq BasicStats Source # 
Instance details

Defined in Algorithm.SRTree.ConfidenceIntervals

data CI Source #

a confience interval is composed of the point estimate (est_), lower bound (_lower_) and upper bound (upper_)

Constructors

CI 

Fields

Instances

Instances details
Read CI Source # 
Instance details

Defined in Algorithm.SRTree.ConfidenceIntervals

Show CI Source # 
Instance details

Defined in Algorithm.SRTree.ConfidenceIntervals

Methods

showsPrec :: Int -> CI -> ShowS #

show :: CI -> String #

showList :: [CI] -> ShowS #

Eq CI Source # 
Instance details

Defined in Algorithm.SRTree.ConfidenceIntervals

Methods

(==) :: CI -> CI -> Bool #

(/=) :: CI -> CI -> Bool #

printCI :: Int -> CI -> IO () Source #

paramCI :: CIType -> Int -> PVector -> Double -> [CI] Source #

Calculates the confidence interval of the parameters using Laplace approximation or Profile likelihood

predictionCI :: CIType -> Distribution -> (SRMatrix -> PVector) -> (SRMatrix -> [PVector]) -> (CI -> PVector -> Fix SRTree -> (Double -> Double, Double)) -> SRMatrix -> Fix SRTree -> PVector -> Double -> [CI] -> [CI] Source #

calculates the prediction confidence interval using Laplace approximation or profile likelihood.