org.lcsim.recon.tracking.trfutil

Class ArraySmearer

java.lang.Object

org.lcsim.recon.tracking.trfutil.ArraySmearer

public class ArraySmearer
extends Object

This class smears a vector of quantities in accordance with a multivariate normal distribution. The constructor takes a covariance matrix as input; only a minimal amount of checking is done. The method 'smear' takes an array and smears its members according to a gaussian distribution, whose width corresponds to the square root of the diagonal element of the covariance matrix. Here is an example of how the covariance terms are handled:

        v11
 cov =  v12 v22
        v13 v23 v33

  x1 = x1 + a1*rn1
  x2 = x2 + a2*rn1 + b2*rn2
  x3 = x3 + a3*rn1 + b3*rn2 + c3*rn3

where rn1, rn2 and rn3 are independently generated normal random numbers

  a1 = sqrt(v11)
  if a1 = 0, then a2 = a3 = 0
  otherwise a2 = v12/a1, a3 = v13/a1

  b2 = sqrt(v22 - a2*a2)
  if b2 = 0, then b3 = 0
  otherwise b3 = (v23 - a2*a3)/b2

  c3 = sqrt(v33 - a3*a3 - b3*b3)

Version:

1.0

Author:

Norman A. Graf

Field Summary

Fields

Modifier and Type	Field and Description
private double[][]	_cov
private int	_ndim
private Random	_r
private double[][]	_smear

Constructor Summary

Constructors

Constructor and Description
ArraySmearer(double[][] cov) constructor from the covariance matrix

Method Summary

Methods

Modifier and Type	Method and Description
void	generate(double[] vec) Generate a vector of variables distributed according to the covariance matrix.
void	smear(double[] vec) Smear a vector according to the covariance matrix.
String	toString() Output Stream

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

_smear

private double[][] _smear

_cov

private double[][] _cov

_ndim

private int _ndim

_r

private Random _r

Constructor Detail

ArraySmearer

public ArraySmearer(double[][] cov)

constructor from the covariance matrix

Parameters:

cov - The covariance matrix as

Method Detail

smear

public void smear(double[] vec)

Smear a vector according to the covariance matrix. Note overwrite of input argument!

Parameters:

vec - The array to be smeared.

generate

public void generate(double[] vec)

Generate a vector of variables distributed according to the covariance matrix.

Parameters:

vec - Array to be filled

toString

public String toString()

Output Stream

Overrides:

toString in class Object

Returns:

String representation of object