Examples of cern.jet.random.engine.MersenneTwister

cern.jet.random.engine.MersenneTwister
fsu.edu/~geo/diehard.html">diehard test of G. Marsaglia and the load test of P. Hellekalek and S. Wegenkittl.
Performance: Its speed is comparable to other modern generators (in particular, as fast as java.util.Random.nextFloat()). 2.5 million calls to raw() per second (Pentium Pro 200 Mhz, JDK 1.2, NT). Be aware, however, that there is a non-negligible amount of overhead required to initialize the data structures used by a MersenneTwister. Code like
```
 double sum = 0.0; for (int i=0; i<100000; ++i) { RandomElement twister = new MersenneTwister(new java.util.Date()); sum += twister.raw(); } 
```
will be wildly inefficient. Consider using
```
 double sum = 0.0; RandomElement twister = new MersenneTwister(new java.util.Date()); for (int i=0; i<100000; ++i) { sum += twister.raw(); } 
```
instead. This allows the cost of constructing the MersenneTwister object to be borne only once, rather than once for each iteration in the loop.
Implementation: After M. Matsumoto and T. Nishimura, "Mersenne Twister: A 623-Dimensionally Equidistributed Uniform Pseudo-Random Number Generator", ACM Transactions on Modeling and Computer Simulation, Vol. 8, No. 1, January 1998, pp 3--30.
More info on Masumoto's homepage.
More info on Pseudo-random number generators is on the Web.
Yet some more info.
The correctness of this implementation has been verified against the published output sequence mt19937-2.out of the C-implementation mt19937-2.c. (Call test(1000) to print the sequence).
Note that this implementation is not synchronized.
Details: MersenneTwister is designed with consideration of the flaws of various existing generators in mind. It is an improved version of TT800, a very successful generator. MersenneTwister is based on linear recurrences modulo 2. Such generators are very fast, have extremely long periods, and appear quite robust. MersenneTwister produces 32-bit numbers, and every k-dimensional vector of such numbers appears the same number of times as k successive values over the period length, for each k <= 623 (except for the zero vector, which appears one time less). If one looks at only the first n <= 16 bits of each number, then the property holds for even larger k, as shown in the following table (taken from the publication cited above):

n

1

2

3

4

5

6

7

8

9

10

11

12 .. 16

17 .. 32

k

19937

9968

6240

4984

3738

3115

2493

2492

1869

1869

1248

1246

623

MersenneTwister generates random numbers in batches of 624 numbers at a time, so the caching and pipelining of modern systems is exploited. The generator is implemented to generate the output by using the fastest arithmetic operations only: 32-bit additions and bit operations (no division, no multiplication, no mod). These operations generate sequences of 32 random bits (int's). long's are formed by concatenating two 32 bit int's. float's are formed by dividing the interval [0.0,1.0] into 2³² sub intervals, then randomly choosing one subinterval. double's are formed by dividing the interval [0.0,1.0] into 2⁶⁴ sub intervals, then randomly choosing one subinterval.
@author wolfgang.hoschek@cern.ch @version 1.0, 09/24/99 @see java.util.Random

    calibrationEngine.addInstrument(swaption, METHOD_SWAPTION_SABR);
    // Calibration
    calibrationEngine.calibrate(curves);
    final HullWhiteOneFactorPiecewiseConstantDataBundle hwBundle = new HullWhiteOneFactorPiecewiseConstantDataBundle(hwParameters, curves);
    // Pricing
    final HullWhiteMonteCarloMethod methodMC = new HullWhiteMonteCarloMethod(new NormalRandomNumberGenerator(0.0, 1.0, new MersenneTwister()), DEFAULT_NB_PATH);
    final CurrencyAmount pvMC = methodMC.presentValue(swaption, swaption.getCurrency(), swaption.getUnderlyingSwap().getFirstLeg().getDiscountCurve(), hwBundle);
    return pvMC.getAmount();
  }

View Full Code Here

    calibrationEngine.addInstrument(calibrationBasket, METHOD_CAP_SABR);
    // Calibration
    calibrationEngine.calibrate(curves);
    final HullWhiteOneFactorPiecewiseConstantDataBundle hwBundle = new HullWhiteOneFactorPiecewiseConstantDataBundle(hwParameters, curves);
    // Pricing
    final HullWhiteMonteCarloMethod methodMC = new HullWhiteMonteCarloMethod(new NormalRandomNumberGenerator(0.0, 1.0, new MersenneTwister()), DEFAULT_NB_PATH);
    final CurrencyAmount pvMC = methodMC.presentValue(annuity, annuity.getCurrency(), annuity.getDiscountCurve(), hwBundle);
    return pvMC.getAmount();
  }

View Full Code Here

  /**
   * Compare explicit formula with Monte-Carlo and long/short and payer/receiver parities.
   */
  public void monteCarlo() {
    HullWhiteMonteCarloMethod methodMC;
    methodMC = new HullWhiteMonteCarloMethod(new NormalRandomNumberGenerator(0.0, 1.0, new MersenneTwister()), 10 * NB_PATH);
    // Seed fixed to the DEFAULT_SEED for testing purposes.
    final MultipleCurrencyAmount pvExplicit = METHOD_HW.presentValue(CAP_LONG, HW_MULTICURVES);
    final MultipleCurrencyAmount pvMC = methodMC.presentValue(CAP_LONG, EUR, HW_MULTICURVES);
    assertEquals("Cap/floor - Hull-White - Monte Carlo", pvExplicit.getAmount(EUR), pvMC.getAmount(EUR), 5.0E+2);
    final double pvMCPreviousRun = 136707.032;
    assertEquals("Swaption physical - Hull-White - Monte Carlo", pvMCPreviousRun, pvMC.getAmount(EUR), TOLERANCE_PV);
    methodMC = new HullWhiteMonteCarloMethod(new NormalRandomNumberGenerator(0.0, 1.0, new MersenneTwister()), 10 * NB_PATH);
    final MultipleCurrencyAmount pvShortMC = methodMC.presentValue(CAP_SHORT, EUR, HW_MULTICURVES);
    assertEquals("Swaption physical - Hull-White - Monte Carlo", -pvMC.getAmount(EUR), pvShortMC.getAmount(EUR), TOLERANCE_PV);
  }

View Full Code Here

  public void performance() {
    long startTime, endTime;
    final MultipleCurrencyAmount pvExplicit = METHOD_HW.presentValue(CAP_LONG, HW_MULTICURVES);
    HullWhiteMonteCarloMethod methodMC;
    final int nbPath = 1000000;
    methodMC = new HullWhiteMonteCarloMethod(new NormalRandomNumberGenerator(0.0, 1.0, new MersenneTwister()), nbPath);
    final int nbTest = 10;
    final double[] pv = new double[nbTest];
    final double[] pvDiff = new double[nbTest];


    startTime = System.currentTimeMillis();

View Full Code Here

  /**
   * Test the present value by approximation vs Monte Carlo.
   */
  public void presentValueMonteCarlo() {
    final int nbPath = 12500;
    final G2ppMonteCarloMethod methodMC = new G2ppMonteCarloMethod(new NormalRandomNumberGenerator(0.0, 1.0, new MersenneTwister()), nbPath);
    final MultipleCurrencyAmount pvMC = methodMC.presentValue(SWAPTION_LONG_PAYER, CUR, G2PP_MULTICURVES);
    final MultipleCurrencyAmount pvApproximation = METHOD_G2PP_APPROXIMATION.presentValue(SWAPTION_LONG_PAYER, G2PP_MULTICURVES);
    assertEquals("Swaption physical - G2++ - present value - approximation vs Monte Carlo", pvApproximation.getAmount(CUR), pvMC.getAmount(CUR), 2.5E+4);
  }

View Full Code Here

    final MultipleCurrencyAmount pvApproximation = METHOD_G2PP_APPROXIMATION.presentValue(SWAPTION_LONG_PAYER, G2PP_MULTICURVES);
    final int[] nbPath = new int[] {12500, 100000, 1000000, 10000000};
    final MultipleCurrencyAmount[] pvMC = new MultipleCurrencyAmount[nbPath.length];
    final double[] pvDiff = new double[nbPath.length];
    for (int loopmc = 0; loopmc < nbPath.length; loopmc++) {
      final G2ppMonteCarloMethod methodMC = new G2ppMonteCarloMethod(new NormalRandomNumberGenerator(0.0, 1.0, new MersenneTwister()), nbPath[loopmc]);
      pvMC[loopmc] = methodMC.presentValue(SWAPTION_LONG_PAYER, CUR, G2PP_MULTICURVES);
      pvDiff[loopmc] = pvApproximation.getAmount(CUR) - pvMC[loopmc].getAmount(CUR);
    }
    assertEquals("Swaption physical - G2++ - present value - approximation vs Monte Carlo", pvApproximation.getAmount(CUR), pvMC[nbPath.length - 1].getAmount(CUR), 1.0E+3);
  }

View Full Code Here

  public void performanceMC() {
    long startTime, endTime;
    final int nbTest = 10;


    final int nbPath = 12500;
    final G2ppMonteCarloMethod methodMC = new G2ppMonteCarloMethod(new NormalRandomNumberGenerator(0.0, 1.0, new MersenneTwister()), nbPath);
    @SuppressWarnings("unused")
    MultipleCurrencyAmount pvMC;


    startTime = System.currentTimeMillis();
    for (int looptest = 0; looptest < nbTest; looptest++) {

View Full Code Here

  /**
   * Compare explicit formula with Monte-Carlo and long/short and payer/receiver parities.
   */
  public void presentValueMonteCarlo() {
    HullWhiteMonteCarloMethod methodMC;
    methodMC = new HullWhiteMonteCarloMethod(new NormalRandomNumberGenerator(0.0, 1.0, new MersenneTwister()), NB_PATH);
    // Seed fixed to the DEFAULT_SEED for testing purposes.
    final MultipleCurrencyAmount pvPayerLongExplicit = METHOD_HW.presentValue(SWAPTION_LONG_PAYER, HW_MULTICURVES);
    final MultipleCurrencyAmount pvPayerLongMC = methodMC.presentValue(SWAPTION_LONG_PAYER, EUR, HW_MULTICURVES);
    assertEquals("Swaption physical - Hull-White - Monte Carlo", pvPayerLongExplicit.getAmount(EUR), pvPayerLongMC.getAmount(EUR), 1.0E+4);
    final double pvMCPreviousRun = 4221400.891;
    assertEquals("Swaption physical - Hull-White - Monte Carlo", pvMCPreviousRun, pvPayerLongMC.getAmount(EUR), TOLERANCE_PV);
    methodMC = new HullWhiteMonteCarloMethod(new NormalRandomNumberGenerator(0.0, 1.0, new MersenneTwister()), NB_PATH);
    final MultipleCurrencyAmount pvPayerShortMC = methodMC.presentValue(SWAPTION_SHORT_PAYER, EUR, HW_MULTICURVES);
    assertEquals("Swaption physical - Hull-White - Monte Carlo", -pvPayerLongMC.getAmount(EUR), pvPayerShortMC.getAmount(EUR), TOLERANCE_PV);
    final MultipleCurrencyAmount pvReceiverLongMC = methodMC.presentValue(SWAPTION_LONG_RECEIVER, EUR, HW_MULTICURVES);
    final MultipleCurrencyAmount pvSwap = SWAP_RECEIVER.accept(PVDC, MULTICURVES);
    assertEquals("Swaption physical - Hull-White - Monte Carlo - payer/receiver/swap parity", pvReceiverLongMC.getAmount(EUR) + pvPayerShortMC.getAmount(EUR), pvSwap.getAmount(EUR), 1.0E+5);

View Full Code Here

   * Tests the curve sensitivity in Monte Carlo approach.
   */
  public void presentValueCurveSensitivityMonteCarlo() {
    final double toleranceDelta = 1.0E+6; // 100 USD by bp
    final MultipleCurrencyMulticurveSensitivity pvcsExplicit = METHOD_HW.presentValueCurveSensitivity(SWAPTION_LONG_PAYER, HW_MULTICURVES).cleaned(TOLERANCE_PV_DELTA);
    final HullWhiteMonteCarloMethod methodMC = new HullWhiteMonteCarloMethod(new NormalRandomNumberGenerator(0.0, 1.0, new MersenneTwister()), NB_PATH);
    final MultipleCurrencyMulticurveSensitivity pvcsMC = methodMC.presentValueCurveSensitivity(SWAPTION_LONG_PAYER, EUR, HW_MULTICURVES).cleaned(TOLERANCE_PV_DELTA);
    AssertSensivityObjects.assertEquals("Swaption physical - Hull-White - presentValueCurveSensitivity - payer/receiver/swap parity", pvcsExplicit, pvcsMC, toleranceDelta);
  }

View Full Code Here

    long startTime, endTime;
    final int nbTest = 25;
    MultipleCurrencyAmount pvMC = MultipleCurrencyAmount.of(EUR, 0.0);
    final MultipleCurrencyMulticurveSensitivity pvcsExplicit = METHOD_HW.presentValueCurveSensitivity(SWAPTION_LONG_PAYER, HW_MULTICURVES);
    MultipleCurrencyMulticurveSensitivity pvcsMC = pvcsExplicit;
    final HullWhiteMonteCarloMethod methodMC = new HullWhiteMonteCarloMethod(new NormalRandomNumberGenerator(0.0, 1.0, new MersenneTwister()), NB_PATH);


    startTime = System.currentTimeMillis();
    for (int looptest = 0; looptest < nbTest; looptest++) {
      pvMC = METHOD_HW_MONTECARLO.presentValue(SWAPTION_LONG_PAYER, EUR, HW_MULTICURVES);
    }

View Full Code Here

0 1 2 3 4 5 6 7 8

TOP

Related Classes of cern.jet.random.engine.MersenneTwister

com.opengamma.analytics.financial.interestrate.annuity.method.AnnuityCouponIborRatchetHullWhiteMethodTest

com.opengamma.analytics.financial.interestrate.annuity.method.AnnuityCouponIborRatchetLMMMethodTest

com.opengamma.analytics.financial.interestrate.annuity.provider.AnnuityCouponIborRatchetHullWhiteMethodTest

com.opengamma.analytics.financial.interestrate.annuity.provider.AnnuityCouponIborRatchetLMMMethodTest

com.opengamma.analytics.financial.interestrate.payments.method.CapFloorIborHullWhiteMethodTest

com.opengamma.analytics.financial.interestrate.payments.method.CapFloorIborLMMDDMethodTest

com.opengamma.analytics.financial.interestrate.payments.provider.CapFloorIborHullWhiteMethodTest

com.opengamma.analytics.financial.interestrate.payments.provider.CapFloorIborLMMDDMethodTest

com.opengamma.analytics.financial.interestrate.PresentValueHullWhiteMonteCarloCalculator

com.opengamma.analytics.financial.interestrate.PresentValueSABRHullWhiteMonteCarloCalculator

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.

n	1	2	3	4	5	6	7	8	9	10	11	12 .. 16	17 .. 32
k	19937	9968	6240	4984	3738	3115	2493	2492	1869	1869	1248	1246	623