KSLCore/ksl.utilities.statistic/JackKnifeEstimator

JackKnifeEstimator

class JackKnifeEstimator(originalData: DoubleArray, estimator: BSEstimatorIfc = BSEstimatorIfc.Average())

Constructors

JackKnifeEstimator

constructor(originalData: DoubleArray, estimator: BSEstimatorIfc = BSEstimatorIfc.Average())

Properties

biasCorrectedJackKnifeEstimate

val biasCorrectedJackKnifeEstimate: Double

var defaultCILevel: Double

jackKnifeBiasEstimate

val jackKnifeBiasEstimate: Double

The estimate is (n-1)x(jne - oe), where n = the number of observations, jne is the jackknife estimate and oe = the original data estimate

jackKnifeEstimate

val jackKnifeEstimate: Double

jackKnifeEstimateOfSE

var jackKnifeEstimateOfSE: Double

jackKnifeReplicates

val jackKnifeReplicates: DoubleArray

Returns a copy of the jack knife replicates. The elements of the array represent the application of the estimator to the original data that has been jack knifed. That is, with the ith observation left out. Thus, replicate0 is the estimated quantity computed from the original data with the 0th element left-out, replicate1 is the estimated quantity with element 1 left-out, etc.

val originalData: DoubleArray

originalDataAverage

val originalDataAverage: Double

originalDataEstimate

var originalDataEstimate: Double

originalDataStatistics

val originalDataStatistics: Statistic

val pseudoValues: DoubleArray

nxoe - (n-1)xjnei, where n is the number of observations, oe= original estimate and jnei is the ith leave one out estimate

val sampleSize: Double

Functions

fun asString(): String

jackKnifeConfidenceInterval

fun jackKnifeConfidenceInterval(level: Double = defaultCILevel): Interval

The c.i. is based on the Student-t distribution and the jackknife estimate and its estimate of the standard error

open override fun toString(): String