Package jmt.engine.dataAnalysis

Source Code of jmt.engine.dataAnalysis.QuantileDataAnalyzer

/**   
  * Copyright (C) 2006, Laboratorio di Valutazione delle Prestazioni - Politecnico di Milano

  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.

  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.

  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */

package jmt.engine.dataAnalysis;

import jmt.engine.dataAnalysis.sorting.HeapSort;
import jmt.engine.dataAnalysis.sorting.SortAlgorithm;
import jmt.engine.math.DoubleArrayList;
import jmt.engine.math.SampleMeanVar;

/**

* Analyzes the data in order to calculate the requested quantiles.
* @author Federico Granata
* Date: 24-lug-2003
* Time: 12.04.49

*/
public class QuantileDataAnalyzer extends DynamicDataAnalyzerImpl {
  //TODO: questa classe funziona anche nel caso di InverseMeasure????
  DoubleArrayList data;

  boolean ordered = false;

  double[] quantile;

  SortAlgorithm sorter;

  /**
   * Creates a QuantileDataAnalyzer.
   * @param  alfa    the quantile required for the confidence interval
   * @param  precision   maximum amplitude of confidence interval
   *                      (precision = maxamplitude / mean)
   * @param maxData  maximum number of data to be analyzed
   *
   * @param quantile Requested quantiles //TODO: giusto??
   * @param sorter Sorting algorithm used to manage data
   */
  public QuantileDataAnalyzer(double alfa, double precision, int maxData, double[] quantile, SortAlgorithm sorter) {
    super(alfa, precision, maxData);
    this.quantile = quantile;
    this.sorter = sorter;
  }

  /**
   * Creates a QuantileDataAnalyzer. A default sorting algorithm is
   * used to manage data.
   * @param  alfa    the quantile required for the confidence interval
   * @param  precision   maximum amplitude of confidence interval
   *                      (precision = maxamplitude / mean)
   * @param maxData  maximum number of data to be analyzed
   *
   * @param quantile Requested quantiles
   */
  public QuantileDataAnalyzer(double alfa, double precision, int maxData, double[] quantile) {
    super(alfa, precision, maxData);
    this.quantile = quantile;
    data = new DoubleArrayList(1024);
    data.add(0);
    sorter = new HeapSort();
  }

  /**
   * Adds the new sample to the statistic.
   * @param newSample the new sample
   * @param Weight the weight of the newSample, if it is not needed put 1.
   * @return true if the confidence interval is smaller than required by
   *          the user, or the data analyzed are too many
   */
  @Override
  public boolean addSample(double newSample, double Weight) {
    if (initialized) {
      data.add(newSample * Weight);
      ordered = false;
    }
    return super.addSample(newSample, Weight);
  }

  /**
   * returns the quantile with the requested probability.
   *
   * @param prob probability of the quantile
   * @return the estiamted quantile
   */
  public double getQuantile(double prob) {
    if (ordered) {
      return data.get((int) (data.getSize() * prob));
    } else {
      sort();
      ordered = true;
      return data.get((int) (data.getSize() * prob));
    }
  }

  /**
   * gets all requested quantiles.
   * @return vector of quantiles.
   */
  public double[] getQuantiles() {
    if (quantile != null) {
      double[] res = new double[quantile.length];
      for (int i = 0; i < res.length; i++) {
        res[i] = getQuantile(quantile[i]);
      }
      return res;
    } else {
      return null;
    }
  }

  /**
   * returns the probability that a number extracted from the empirical
   * distibution analyzed is greater then the quantile.
   * @param quantile the requested quantile
   * @return estiamted probability
   */
  public double getProbability(double quantile) {
    if (ordered) {
      return search(quantile);
    }
    return Double.NaN;
  }

  protected double search(double element) {
    int l = 1, r = data.getSize() - 1, x;
    while (l < r) {
      x = (l + r) >> 1;
      if (element == data.get(x)) {
        return (x / (data.getSize() - 1.0));
      }
      if (element < data.get(x)) {
        r = x - 1;
      } else {
        l = x + 1;
      }
    }
    return (l / (data.getSize() - 1.0));
  }

  protected void sort() {
    double[] d = data.toArray(0, data.getSize() - 1);
    //    long start = System.currentTimeMillis();
    sorter.sort(d);
    //    System.out.println("tempo = "+ (System.currentTimeMillis() - start));
    data = new DoubleArrayList(d);

  }

  /** Applies the spetctral test to generate the Confidence Intervals.
   *  see: P. Heidelberger, Peter D. Welch
   * "A spectral method for confidence interval generation and run length
   * control in simulations"
   *
   *
   * @return true if the precision requirement met. false if not.
   */
  @Override
  protected boolean HWtest() {
    sort();
    ordered = true;
    //    System.out.println("quantile " + getQuantileResults(0.75));
    //    System.out.println("nSamples = " + nSamples);
    return super.HWtest();
  }

  /**
   *updates the variance
   */
  @Override
  protected void calcVar() {
    double[] C;
    double[] tempBatch = new double[numBatch];
    double sampleVar = (new SampleMeanVar(batchMean)).getVar();
    K = numBatch / 4;
    C = calcConstants(K, polyOrder);
    C1 = C[0];
    C2 = (int) C[1];
    //DEK (Federico Granata)
    //si puo' fare array copy
    System.arraycopy(batchMean, 0, tempBatch, 0, numBatch - 1);
    //    for (int i = 0; i < numBatch; i++)
    //      tempBatch[i] = batchMean[i];

    extVar = calcVar(tempBatch, 0, batch, K, polyOrder);
    if (Math.abs(extVar - sampleVar) > sampleVar * precision * 2) {
      extVar = Double.MAX_VALUE;
    }
    if (extVar < sampleVar) {
      extVar = sampleVar;
    }
  }

}
TOP

Related Classes of jmt.engine.dataAnalysis.QuantileDataAnalyzer

TOP
Copyright © 2018 www.massapi.com. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.