RunningWeightedStatistics - Math.NET Numerics Documentation

Namespaces

Types in MathNet.Numerics.Statistics

Running weighted statistics accumulator, allows updating by adding values or by combining two accumulators. Weights are reliability weights, not frequency weights.

This type declares a DataContract for out of the box ephemeral serialization with engines like DataContractSerializer, Protocol Buffers and FsPickler, but does not guarantee any compatibility between versions. It is not recommended to rely on this mechanism for durable persistence.

Constructors

Static Functions

Combine

Methods

Properties

Public Constructors

RunningWeightedStatistics()

RunningWeightedStatistics(IEnumerable<Tuple<double, double>> values)

Public Static Functions

RunningWeightedStatistics Combine(RunningWeightedStatistics a, RunningWeightedStatistics b)

Create a new running statistics over the combined samples of two existing running statistics.

Public Methods

bool Equals(object obj)

int GetHashCode()

Type GetType()

void Push(double weight, double value)

Update the running statistics by adding another observed sample (in-place).

void PushRange(IEnumerable<Tuple<double, double>> values)

Update the running statistics by adding a sequence of weighted observatopms (in-place).

void PushRange(IEnumerable<double> weights, IEnumerable<double> values)

Update the running statistics by adding a sequence of weighted observatopms (in-place).

string ToString()

Public Properties

long Count get;

Gets the total number of samples with non-zero weight.

double EffectiveSampleSize get;

The Kish's Effective Sample Size

double Kurtosis get;

Estimates the unbiased population excess kurtosis from the provided samples. Will use the Bessel correction for reliability weighting. Returns NaN if data has less than four entries or if any entry is NaN. Equivalent formula for this for weighted distributions are unknown.

double Maximum get;

Returns the maximum value in the sample data. Returns NaN if data is empty or if any entry is NaN.

double Mean get;

Evaluates the sample mean, an estimate of the population mean. Returns NaN if data is empty or if any entry is NaN.

double Minimum get;

Returns the minimum value in the sample data. Returns NaN if data is empty or if any entry is NaN.

double PopulationKurtosis get;

Evaluates the population excess kurtosis from the full population. Does not use a normalizer and would thus be biased if applied to a subset (type 1). Returns NaN if data has less than three entries or if any entry is NaN.

double PopulationSkewness get;

Evaluates the population skewness from the full population. Does not use a normalizer and would thus be biased if applied to a subset (type 1). Returns NaN if data has less than two entries or if any entry is NaN.

double PopulationStandardDeviation get;

Evaluates the standard deviation from the provided full population. On a dataset of size N will use an N normalizer and would thus be biased if applied to a subset. Returns NaN if data is empty or if any entry is NaN.

double PopulationVariance get;

Evaluates the variance from the provided full population. On a dataset of size N will use an N normalizer and would thus be biased if applied to a subset. Returns NaN if data is empty or if any entry is NaN.

double Skewness get;

Estimates the unbiased population skewness from the provided samples. Will use the Bessel correction for reliability weighting. Returns NaN if data has less than three entries or if any entry is NaN.

double StandardDeviation get;

Estimates the unbiased population standard deviation from the provided samples. Will use the Bessel correction for reliability weighting. Returns NaN if data has less than two entries or if any entry is NaN.

double TotalWeight get;

Evaluates the total weight of the population.

double Variance get;

Estimates the unbiased population variance from the provided samples. Will use the Bessel correction for reliability weighting. Returns NaN if data has less than two entries or if any entry is NaN.