public void CanEstimateParameters(double shape, double scale) { var original = new Weibull(shape, scale, new Random(100)); var estimated = Weibull.Estimate(original.Samples().Take(10000)); AssertHelpers.AlmostEqualRelative(shape, estimated.Shape, 1); AssertHelpers.AlmostEqualRelative(scale, estimated.Scale, 1); }
public void CanSampleSequenceStatic() { var ied = Weibull.Samples(new Random(0), 1.0, 1.0); GC.KeepAlive(ied.Take(5).ToArray()); }
public void CanSampleSequenceStatic() { var ied = Weibull.Samples(new Random(), 1.0, 1.0); var arr = ied.Take(5).ToArray(); }
public double[] GetSampleData(string distType, double mostLikelyEstimate, double lowEstimate, double highEstimate) { if (Iterations > 10000) { Iterations = 10000; } if (Iterations <= 2) { Iterations = 1000; } if (this.CILevel < 10) { this.CILevel = 90; } if (this.CILevel > 99) { this.CILevel = 99; } Random rnd = new Random(Random); mostLikelyEstimate = Math.Round(mostLikelyEstimate, 4); lowEstimate = Math.Round(lowEstimate, 4); highEstimate = Math.Round(highEstimate, 4); var sampledata = new double[Iterations]; if (distType == Calculator1.RUC_TYPES.triangle.ToString()) { if (lowEstimate >= mostLikelyEstimate || lowEstimate == 0) { //arbitrary rules (25%) lowEstimate = mostLikelyEstimate * .75; //no errors: lowEstimate = 0 is often the case //sb.AppendLine(Errors.GetMessage("DATA_BADDISTRIBUTION")); } if (highEstimate <= mostLikelyEstimate || highEstimate == 0) { //arbitrary rules (25%) highEstimate = mostLikelyEstimate * 1.25; } if (Random != 0) { //generate samples of the Triangular(low, high, mode) distribution; Triangular.Samples(rnd, sampledata, lowEstimate, highEstimate, mostLikelyEstimate); } else { //generate samples of the Triangular(low, high, mode) distribution; Triangular.Samples(sampledata, lowEstimate, highEstimate, mostLikelyEstimate); } } else if (distType == Calculator1.RUC_TYPES.normal.ToString()) { //generate samples of the Normal(mean, sd) distribution; if (Random != 0) { Normal.Samples(rnd, sampledata, lowEstimate, highEstimate); } else { Normal.Samples(sampledata, lowEstimate, highEstimate); } } else if (distType == Calculator1.RUC_TYPES.lognormal.ToString()) { if (Random != 0) { LogNormal.Samples(rnd, sampledata, lowEstimate, highEstimate); } else { LogNormal.Samples(sampledata, lowEstimate, highEstimate); } } else if (distType == Calculator1.RUC_TYPES.weibull.ToString()) { if (Random != 0) { Weibull.Samples(rnd, sampledata, lowEstimate, highEstimate); } else { Weibull.Samples(sampledata, lowEstimate, highEstimate); } } else if (distType == Calculator1.RUC_TYPES.beta.ToString()) { if (Random != 0) { Beta.Samples(rnd, sampledata, lowEstimate, highEstimate); } else { Beta.Samples(sampledata, lowEstimate, highEstimate); } } else if (distType == Calculator1.RUC_TYPES.pareto.ToString()) { if (Random != 0) { Pareto.Samples(rnd, sampledata, lowEstimate, highEstimate); } else { Pareto.Samples(sampledata, lowEstimate, highEstimate); } } else if (distType == Calculator1.RUC_TYPES.uniform.ToString()) { var sampleints = new int[Iterations]; int iLower = CalculatorHelpers.ConvertStringToInt(lowEstimate.ToString()); int iUpper = CalculatorHelpers.ConvertStringToInt(highEstimate.ToString()); if (Random != 0) { DiscreteUniform.Samples(rnd, sampleints, iLower, iUpper); } else { DiscreteUniform.Samples(sampleints, iLower, iUpper); } for (int i = 0; i < sampleints.Count(); i++) { sampledata[i] = sampleints[i]; } } else if (distType == Calculator1.RUC_TYPES.bernoulli.ToString()) { var sampleints = new int[Iterations]; if (Random != 0) { Bernoulli.Samples(rnd, sampleints, lowEstimate); } else { Bernoulli.Samples(sampleints, lowEstimate); } for (int i = 0; i < sampleints.Count(); i++) { sampledata[i] = sampleints[i]; } } else if (distType == Calculator1.RUC_TYPES.poisson.ToString()) { var sampleints = new int[Iterations]; if (Random != 0) { Poisson.Samples(rnd, sampleints, lowEstimate); } else { Poisson.Samples(sampleints, lowEstimate); } for (int i = 0; i < sampleints.Count(); i++) { sampledata[i] = sampleints[i]; } } else if (distType == Calculator1.RUC_TYPES.binomial.ToString()) { var sampleints = new int[Iterations]; int iUpperEstimate = CalculatorHelpers.ConvertStringToInt(highEstimate.ToString()); if (Random != 0) { Binomial.Samples(rnd, sampleints, lowEstimate, iUpperEstimate); } else { Binomial.Samples(sampleints, lowEstimate, iUpperEstimate); } for (int i = 0; i < sampleints.Count(); i++) { sampledata[i] = sampleints[i]; } } else if (distType == Calculator1.RUC_TYPES.gamma.ToString()) { //generate samples of the Gamma(shape, scale) distribution; if (Random != 0) { Gamma.Samples(rnd, sampledata, lowEstimate, highEstimate); } else { Gamma.Samples(sampledata, lowEstimate, highEstimate); } } else { //don't force them to use distribution } //hold for possible infernet use //else if (distType == Calculator1.RUC_TYPES.dirichlet.ToString()) //{ // //generate samples of the Dirichlet(random, alpha) distribution; // Dirichlet.Sample(sampledata, lowEstimate); //} //else if (distType == Calculator1.RUC_TYPES.wishart.ToString()) //{ // //generate samples of the Wishart(random, degrees of freedom, scale) distribution; // Wishart.Sample(sampledata, lowEstimate, highEstimate); //} //the mathlibrary supports more than a dozen additional distributions return(sampledata); }
/// <summary> /// Run example /// </summary> /// <a href="http://en.wikipedia.org/wiki/Weibull_distribution">Weibull distribution</a> public void Run() { // 1. Initialize the new instance of the Weibull distribution class with parameters Scale = 1, Shape = 0.5 var weibull = new Weibull(0.5, 1); Console.WriteLine(@"1. Initialize the new instance of the Weibull distribution class with parameterы Scale = {0}, Shape = {1}", weibull.Scale, weibull.Shape); Console.WriteLine(); // 2. Distributuion properties: Console.WriteLine(@"2. {0} distributuion properties:", weibull); // Cumulative distribution function Console.WriteLine(@"{0} - Сumulative distribution at location '0.3'", weibull.CumulativeDistribution(0.3).ToString(" #0.00000;-#0.00000")); // Probability density Console.WriteLine(@"{0} - Probability density at location '0.3'", weibull.Density(0.3).ToString(" #0.00000;-#0.00000")); // Log probability density Console.WriteLine(@"{0} - Log probability density at location '0.3'", weibull.DensityLn(0.3).ToString(" #0.00000;-#0.00000")); // Entropy Console.WriteLine(@"{0} - Entropy", weibull.Entropy.ToString(" #0.00000;-#0.00000")); // Largest element in the domain Console.WriteLine(@"{0} - Largest element in the domain", weibull.Maximum.ToString(" #0.00000;-#0.00000")); // Smallest element in the domain Console.WriteLine(@"{0} - Smallest element in the domain", weibull.Minimum.ToString(" #0.00000;-#0.00000")); // Mean Console.WriteLine(@"{0} - Mean", weibull.Mean.ToString(" #0.00000;-#0.00000")); // Median Console.WriteLine(@"{0} - Median", weibull.Median.ToString(" #0.00000;-#0.00000")); // Mode Console.WriteLine(@"{0} - Mode", weibull.Mode.ToString(" #0.00000;-#0.00000")); // Variance Console.WriteLine(@"{0} - Variance", weibull.Variance.ToString(" #0.00000;-#0.00000")); // Standard deviation Console.WriteLine(@"{0} - Standard deviation", weibull.StdDev.ToString(" #0.00000;-#0.00000")); // Skewness Console.WriteLine(@"{0} - Skewness", weibull.Skewness.ToString(" #0.00000;-#0.00000")); Console.WriteLine(); // 3. Generate 10 samples of the Weibull distribution Console.WriteLine(@"3. Generate 10 samples of the Weibull distribution"); for (var i = 0; i < 10; i++) { Console.Write(weibull.Sample().ToString("N05") + @" "); } Console.WriteLine(); Console.WriteLine(); // 4. Generate 100000 samples of the Weibull(0.5, 1) distribution and display histogram Console.WriteLine(@"4. Generate 100000 samples of the Weibull(0.5, 1) distribution and display histogram"); var data = new double[100000]; Weibull.Samples(data, 0.5, 1.0); ConsoleHelper.DisplayHistogram(data); Console.WriteLine(); // 5. Generate 100000 samples of the Weibull(1.5, 1) distribution and display histogram Console.WriteLine(@"5. Generate 100000 samples of the Weibull(1.5, 1) distribution and display histogram"); Weibull.Samples(data, 1.5, 1.0); ConsoleHelper.DisplayHistogram(data); Console.WriteLine(); // 6. Generate 100000 samples of the Weibull(5, 1) distribution and display histogram Console.WriteLine(@"6. Generate 100000 samples of the Weibull(5, 1) distribution and display histogram"); Weibull.Samples(data, 5.0, 1.0); ConsoleHelper.DisplayHistogram(data); }