Companion

object Companion

Properties

Link copied to clipboard

This set contains all the known estimators for estimating continuous distributions. This is the union of nonRestrictedEstimators and positiveRestrictedEstimators

Link copied to clipboard

How close we consider a double is to 0.0 to consider it 0.0 Default is 0.95

Link copied to clipboard

For rank based evaluation, this specifies the default parameter value for those methods the perform rank based evaluation calculations.

Link copied to clipboard

This set holds predefined scoring models for evaluating the distribution goodness of fit.

Link copied to clipboard

How close we consider a double is to 0.0 to consider it 0.0 Default is 0.001

Link copied to clipboard

This set holds estimators that can fit distributions for which the domain is not restricted.

Link copied to clipboard

This set holds the recommended estimators for estimating the parameters of distributions on the positive domain x in (0, infinity)

Functions

Link copied to clipboard

Returns the probability of being in each bin, F(upper limit) - F(lower limit) within the bins with p0 for bins0 etc.

Link copied to clipboard

Returns the probability of being in the bin, F(upper limit) - F(lower limit)

Link copied to clipboard
fun bootstrapFamilyFrequency(data: DoubleArray, evaluationMethod: EvaluationMethod = EvaluationMethod.Scoring, estimators: Set<ParameterEstimatorIfc> = allEstimators, scoringModels: Set<PDFScoringModel> = defaultScoringModels, numBootstrapSamples: Int = 400, automaticShifting: Boolean = true, stream: RNStreamIfc = KSLRandom.nextRNStream()): IntegerFrequency

Constructs a frequency tabulation of the top distributions for each bootstrap sample of the original data. This function can be used to explore the set of distributions that may best-fit the data.

Link copied to clipboard
fun bootstrapFamilyFrequencyAsDataFrame(data: DoubleArray, evaluationMethod: EvaluationMethod = EvaluationMethod.Scoring, estimators: Set<ParameterEstimatorIfc> = allEstimators, scoringModels: Set<PDFScoringModel> = defaultScoringModels, numBootstrapSamples: Int = 400, automaticShifting: Boolean = true, stream: RNStreamIfc = KSLRandom.nextRNStream()): AnyFrame

Constructs a frequency tabulation of the top distributions for each bootstrap sample of the original data. This function can be used to explore the set of distributions that may best-fit the data.

Link copied to clipboard
fun confidenceIntervalForMaximum(data: DoubleArray, numBootstrapSamples: Int = 399, level: Double = 0.95): Interval

Uses bootstrapping to estimate a confidence interval for the maximum

Link copied to clipboard
fun confidenceIntervalForMinimum(data: DoubleArray, numBootstrapSamples: Int = 399, level: Double = 0.95): Interval

Uses bootstrapping to estimate a confidence interval for the minimum

Link copied to clipboard
fun createDefaultPDFEvaluationModel(scoringResults: List<ScoringResult>, rankingMethod: Statistic.Companion.Ranking = defaultRankingMethod): AdditiveMODAModel

Creates an additive evaluation model based on the metrics within the scoring results that has linear value functions for the metrics. The list of scoring results must not be empty.

Link copied to clipboard

Constructs an instance of the appropriate continuous probability distribution for the provided random variable parameters. If no probability distribution is defined for the supplied type of random variable, then null is returned.

Link copied to clipboard

Computes breakpoints for the distribution that ensures (approximately) that the expected number of observations within the intervals defined by the breakpoints will be equal. That is, the probability associated with each interval is equal. In addition, the expected number of observations will be approximately greater than or equal to 5. There will be at least two breakpoints and thus at least 3 intervals defined by the breakpoints.

Link copied to clipboard
fun estimateLeftShift(min: Double, nextSmallest: Double, max: Double, tolerance: Double = defaultZeroTolerance): Double

Uses the method described on page 360 of Law (2007) Simulation Modeling and Analysis, ISBN 0073294411, 9780073294414 The min must be strictly less than the max. The nextSmallest is x(k) where x(k) is the kth order statistic and k is the value in {2, 3, ..., n-1} such that x(k) is strictly greater than x(1).

Link copied to clipboard
fun estimateLeftShiftParameter(data: DoubleArray, tolerance: Double = defaultZeroTolerance): Double

Uses the method described on page 360 of Law (2007) Simulation Modeling and Analysis, ISBN 0073294411, 9780073294414 There must be at least three observations within the data and there must be at least three different values. That is, all the values must not be the same. The observations should all be greater than or equal to 0.0. That is, no negative values are allowed within the data. If these conditions do not hold, then 0.0 is returned for the shift. That is, no shift.

Link copied to clipboard

The expected number of observations within the bin given a particular cdf

Link copied to clipboard

The expected number of observations in each bin given a particular cdf.

Link copied to clipboard

If min is the minimum of the data, then this computes the observation that is strictly larger than the observed minimum. This is the minimum of the observations if all observations of min are removed from the data. There must be at least 2 observations and the observations cannot all be equal. There must be at least one observation that is not equal to min.

Link copied to clipboard
fun leftShiftData(data: DoubleArray, tolerance: Double = defaultZeroTolerance): ShiftedData

Estimates the shift parameter and then shifts the data by the estimated quantity. Also returns the computed shift. Use destructuring if you want:

Link copied to clipboard
fun rangeEstimate(min: Double, max: Double, n: Int): Interval

Estimates the range based on uniform distribution theory. Uses the minimum unbiased estimators based on the order statistics. See: 1. Castillo E, Hadi AS. A method for estimating parameters and quantiles of distributions of continuous random variables. Computational Statistics & Data Analysis. 1995 Oct;20(4):421–39.