accuracyindicators.java

搞算法预测的可以来看。有移动平均法

//
//  OpenForecast - open source, general-purpose forecasting package.
//  Copyright (C) 2002  Steven R. Gould
//
//  This library is free software; you can redistribute it and/or
//  modify it under the terms of the GNU Lesser General Public
//  License as published by the Free Software Foundation; either
//  version 2.1 of the License, or (at your option) any later version.
//
//  This library 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
//  Lesser General Public License for more details.
//
//  You should have received a copy of the GNU Lesser General Public
//  License along with this library; if not, write to the Free Software
//  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
//

package net.sourceforge.openforecast.models;


/**
 * A simple Java Bean that gathers together in one class all current "accuracy
 * indicators" for a given forecasting model. Currently, the supported
 * measurements of accuracy of a forecasting model supported by this class
 * include:
 * <ul>
 *  <li>bias - or the mean error</li>
 *  <li>MAD - or the Mean Absolute Deviation</li>
 *  <li>MAPE - or the Mean Absolute Percentage Error</li>
 *  <li>MSE - or the Mean Square of the Error</li>
 *  <li>SAE - or the Sum of the Absolute Errors</li>
 * </ul>
 * @since 0.3
 * @author Steven R. Gould
 */
class AccuracyIndicators
{
	/** Arithmetic mean of the errors. */
	private double bias;
	
	/** Mean Absolute Deviation. */
	private double mad;

	/** Mean Absolute Percentage Error. */
	private double mape;

	/** Mean Square of the Error. */
	private double mse;

	/** Sum of the Absolute Errors. */
	private double sae;

	/**
	 * Default constructor. Initializes all accuracy indicators to their
	 * "worst" possible values - generally this means some large number,
	 * indicating "very" inaccurate.
	 */
	public AccuracyIndicators()
	{
		bias = mad = mape = mse = sae = Double.MAX_VALUE;
	}

	/**
	 * Returns the bias for the associated forecasting model.
	 * @return the bias.
	 */
	public double getBias()
	{
		return bias;
	}

	/**
	 * Sets the bias for the associated forecasting model to the given value.
	 * @param bias the new value for the bias.
	 */
	public void setBias(double bias)
	{
		this.bias = bias;
	}

	/**
	 * Returns the Mean Absolute Deviation (MAD) for the associated
	 * forecasting model to the given value.
	 * @return the Mean Absolute Deviation.
	 */
	public double getMAD()
	{
		return mad;
	}

	/**
	 * Sets the Mean Absolute Deviation (MAD) for the associated forecasting
	 * model. That is, the <code>SUM( abs(actual-forecast) ) / n</code> for
	 * the initial data set.
	 * @param mad the new value for the Mean Absolute Deviation.
	 */
	public void setMAD(double mad)
	{
		this.mad = mad;
	}

	/**
	 * Returns the Mean Absolute Percentage Error (MAPE) for the associated
	 * forecasting model. That is, the
	 * <code>SUM( 100% . abs(actual-forecast)/actual ) / n</code> for the
	 * initial data set.
	 * @return the Mean Absolute Percentage Error.
	 */
	public double getMAPE()
	{
		return mape;
	}

	/**
	 * Sets the Mean Absolute Percentage Error (MAPE) for the associated
	 * forecasting model to the given value.
	 * @param mape the new value for the Mean Absolute Percentage Error.
	 */
	public void setMAPE(double mape)
	{
		this.mape = mape;
	}

	/**
	 * Returns the Mean Square of the Errors (MSE) for the associated
	 * forecasting model. That is, the
	 * <code>SUM( (actual-forecast)^2 ) / n</code> for the initial data set.
	 * @return the Mean Square of the Errors.
	 */
	public double getMSE()
	{
		return mse;
	}

	/**
	 * Sets the Mean Square of the Errors (MSE) for the associated
	 * forecasting model to the new value.
	 * @param mse the new value for the Mean Square of the Errors.
	 */
	public void setMSE(double mse)
	{
		this.mse = mse;
	}

	/**
	 * Returns the Sum of the Absolute Errors (SAE) for the associated
	 * forecasting model. That is, the
	 * <code>SUM( abs(actual-forecast) )</code> for the initial data set.
	 * @return the Mean Absolute Deviation.
	 */
	public double getSAE()
	{
		return sae;
	}

	/**
	 * Sets the Sum of the Absolute Errors (SAE) for the associated
	 * forecasting model to the new value.
	 * @return the new value for the Mean Absolute Deviation.
	 */
	public void setSAE(double sae)
	{
		this.sae = sae;
	}

	/**
	 * Returns a string containing the accuracy indicators and their values.
	 * Overridden to provide some useful output for debugging.
	 * @return a string containing the accuracy indicators and their values.
	 */
	public String toString()
	{
		return "bias=" + bias
			+ ", MAD=" + mad
			+ ", MAPE=" + mape
			+ ", MSE=" + mse
			+ ", sae=" + sae;
	}
}