linear_algebra
Class QRTest_j

java.lang.Object
  |
  +--linear_algebra.QRTest_j

public class QRTest_j
extends java.lang.Object

This class tests the (LINPACK) QR classes. It

  1. Randomly fills an n by n matrix, X.
  2. Randomly generates a vector b0.
  3. Calculates the vector y = X(b0).
  4. Performs a QR decomposition of X.
  5. Solves the system Xb = y in an effort to recover b0.
  6. Handles the regression equation y = a0 + a1*x + a2*x^2 + e where a0 = a1 = a2 = 1. Its estimates of the parameters should be close to 1. Its estimate of the standard deviation should be close to the input value.


Constructor Summary
QRTest_j()
           
 
Method Summary
static void isort(int[] x, int[] xsort, int[] origord, int n)
           
static void main(java.lang.String[] args)
           
static double normi(double u)
           This is a normal cdf inverse routine.
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

QRTest_j

public QRTest_j()
Method Detail

main

public static void main(java.lang.String[] args)

normi

public static double normi(double u)

This is a normal cdf inverse routine. Created by Steve Verrill, March 1997.

Parameters:
u - The value (between 0 and 1) to invert.

isort

public static void isort(int[] x,
                         int[] xsort,
                         int[] origord,
                         int n)