KSLCore/ksl.utilities.distributions.fitting.estimators/WeibullPercentileParameterEstimator

WeibullPercentileParameterEstimator

class WeibullPercentileParameterEstimator(name: String? = "WeibullPercentileParameterEstimator") : ParameterEstimatorIfc, MVBSEstimatorIfc, IdentityIfc

Applies the percentile method to estimating the parameters of the Weibull distribution as outline in these papers:

Marks NB. Estimation of Weibull parameters from common percentiles. Journal of Applied Statistics. 2005 Jan;32(1):17–24.

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.

The strategy is to apply the method described in Marks (2005) using multiple pairings of complementary percentiles forming a sample of shape estimates as noted in Castillo and Hadi (1995).

We then apply a jack knife estimation procedure on the generated shape estimates to form the final estimated shape. Once the shape is estimated, the scale parameter is estimated using the recommended method described on pages 287-288 of Law (2007) Simulation Modeling and Analysis.

The procedure requires that at least 10 observations are available. The user can adjust the percentiles used through the fields reducedPercentileSet or expandedPercentileSet. Alternative, the user can directly provide the lower half of the percentile set via the property percentileSet. After executing the estimation process on some data, the jackKnifeEstimator field can be used to access the computed shape estimates.

The supplied data cannot be negative or zero and must not all be equal in value.

Constructors

WeibullPercentileParameterEstimator

constructor(name: String? = "WeibullPercentileParameterEstimator")

Properties

checkRange

open override val checkRange: Boolean = true

Indicates if the estimator requires that the range of the data be checked for a shift before the estimation process.

expandedPercentileSet

val expandedPercentileSet: DoubleArray

open override val id: Int

jackKnifeEstimator

var jackKnifeEstimator: JackKnifeEstimator?

Holds the estimated shape parameters from the percentile method. This field will be overwritten during the estimation process.

label

open override var label: String?

name

open override val name: String

names

open override val names: List<String>

The name to associate with each dimension of the array that is returned by estimate(). The names should be unique. The order of the list of names should match the order of elements in the returned array.

percentileSet

var percentileSet: DoubleArray

Use to specify the set of percentiles that will be used during the estimation process. This is the lower half of the percentiles. These will be matched with their complement values to form the pairs used to estimate shape and scale. For example 0.1 and 0.9 are a common pair used.

reducedPercentileSet

val reducedPercentileSet: DoubleArray

rvType

open override val rvType: RVParametersTypeIfc

The type of random variable for which this estimator estimates parameters.

sampleSizeFactor

var sampleSizeFactor: Int

If the size of the sample is less than this value, then use a reduced percentile set

Functions

estimate

open override fun estimate(data: DoubleArray): DoubleArray

If the estimation process is not successful, then an empty array is returned.

estimateParameters

open override fun estimateParameters(data: DoubleArray, statistics: StatisticIfc): EstimationResult

Estimates the parameters associated with some distribution. The returned EstimationResult needs to be consistent with the intent of the desired distribution. Note the meaning of the fields associated with EstimationResult