Diffpack Documentation


Main Page   Class Hierarchy   Alphabetical List   Compound List   File List   Compound Members   File Members   Search  

ExpecVarFieldLattice Class Reference

estimation of expectation and variance of fields. More...

#include <ExpecVarFieldLattice.h>

Inheritance diagram for ExpecVarFieldLattice::

EmpiricalStatistics MomentsFieldLattice List of all members.

Public Methods

 ExpecVarFieldLattice ()
 ~ExpecVarFieldLattice ()
bool redim (const FieldLattice &f)
bool ok () const
void init ()
void update (FieldLattice &u)
FieldLatticegetExpec ()
FieldLatticegetStDev ()
FieldLatticegetVar ()

Protected Methods

void estimator ()

Protected Attributes

Handle(FieldLatticez1
Handle(FieldLatticez2
Handle(FieldLatticeEU
Handle(FieldLatticeSDU
Handle(FieldLatticeVARU

Detailed Description

estimation of expectation and variance of fields.

NAME: ExpecVarFieldLattice - estimation of expectation and variance of fields

DESCRIPTION:

The class offers estimation of the mean value and the variance of a stochastic field over a lattice grid (typically a finite difference-type of field). The methods are in principle similar to class "ExpecVar", but instead of estimating random variables, one estimates random fields. There is no estimation of the spatial correlation structure of the field, only pointwise estimation. The "CorrAlongLine" class can be used to estimate correlation structures.


Constructor & Destructor Documentation

ExpecVarFieldLattice::ExpecVarFieldLattice ( )
 

The constructor takes no arguments. For initialization it is in addition required to call the "redim" function. The parameter is a "FieldLattice" field which is used to redimension the various internal "FieldLattice" objects that are needed for collecting the estimated quantities.

Prior to collecting samples of fields one has to call "init", which initializes the statistics for the mean and variance of the field.

ExpecVarFieldLattice::~ExpecVarFieldLattice ( ) [inline]
 


Member Function Documentation

void ExpecVarFieldLattice::estimator ( ) [protected]
 

Reimplemented in MomentsFieldLattice.

FieldLattice & ExpecVarFieldLattice::getExpec ( )
 

returns a reference to the estimated expecation field (so far). Note that the expectation field is administered by this class (allocated, deallocated, computed, stored).

FieldLattice & ExpecVarFieldLattice::getStDev ( )
 

same as "getExpec" but it concerns the standard deviation field.

FieldLattice & ExpecVarFieldLattice::getVar ( )
 

same as "getExpec" but it concerns the variance field.

void ExpecVarFieldLattice::init ( )
 

initializes the values of the internal data structures used for the estimation. In other words, "redim" alloocates the data structures, while "init" sets the field entries to zero (initializing the statistics for the mean and variance). It is usually called from an "initEstimator" function in the simulator class.

Reimplemented from EmpiricalStatistics.

Reimplemented in MomentsFieldLattice.

bool ExpecVarFieldLattice::ok ( ) const
 

Reimplemented in MomentsFieldLattice.

bool ExpecVarFieldLattice::redim ( const FieldLattice & f )
 

redimensions the internal data structures so that they correspond in size to the field given as argument to the function.

Reimplemented in MomentsFieldLattice.

void ExpecVarFieldLattice::update ( FieldLattice & u )
 

adds a sample of a random field to the internal statistics for the mean and variance estimation. For example, the sample field is just added to the estimator for the expectation, and the square of the field values are added to the estimator for the variance.

Reimplemented in MomentsFieldLattice.


Member Data Documentation

Handle(FieldLattice) ExpecVarFieldLattice::EU [protected]
 

Handle(FieldLattice) ExpecVarFieldLattice::SDU [protected]
 

Handle(FieldLattice) ExpecVarFieldLattice::VARU [protected]
 

Handle(FieldLattice) ExpecVarFieldLattice::z1 [protected]
 

Handle(FieldLattice) ExpecVarFieldLattice::z2 [protected]
 


The documentation for this class was generated from the following file:
Copyright © 2003 inuTech GmbH. All rights reserved.