UnivariateStatisticsUtil (Cognitive Foundry)

java.lang.Object
- gov.sandia.cognition.math.UnivariateStatisticsUtil

```
public class UnivariateStatisticsUtil
extends java.lang.Object
```
Some static methods for computing generally useful univariate statistics.

Since:

2.0

Author:

Kevin R. Dixon

Constructor Summary

Constructors
Constructor and Description

UnivariateStatisticsUtil()

Constructors
Constructor and Description
`UnivariateStatisticsUtil()`

Method Summary

All Methods Static Methods Concrete Methods
Modifier and Type	Method and Description
`static double`	`computeCentralMoment(java.lang.Iterable<? extends java.lang.Number> data, double mean, int moment)` Computes the desired biased estimate central moment of the given dataset.
`static double`	`computeCorrelation(java.util.Collection<? extends java.lang.Number> data1, java.util.Collection<? extends java.lang.Number> data2)` Computes the correlation coefficient in a single pass.
`static double`	`computeEntropy(java.lang.Iterable<? extends java.lang.Number> data)` Computes the information-theoretic entropy of the PMF in bits (base 2).
`static double`	`computeKurtosis(java.util.Collection<? extends java.lang.Number> data)` Computes the biased excess kurtosis of the given dataset.
`static double`	`computeMaximum(java.lang.Iterable<? extends java.lang.Number> data)` Finds the maximum value of a data set.
`static double`	`computeMean(java.lang.Iterable<? extends java.lang.Number> data)` Computes the arithmetic mean (average, expectation, first central moment) of a dataset
`static Pair<java.lang.Double,java.lang.Double>`	`computeMeanAndVariance(java.lang.Iterable<? extends java.lang.Number> data)` Computes the mean and unbiased variance of a Collection of data using the one-pass approach.
`static double`	`computeMedian(java.util.Collection<? extends java.lang.Number> data)` Computes the median of the given data.
`static Pair<java.lang.Double,java.lang.Double>`	`computeMinAndMax(java.lang.Iterable<? extends java.lang.Number> data)` Computes the minimum and maximum of a set of data in a single pass.
`static double`	`computeMinimum(java.lang.Iterable<? extends java.lang.Number> data)` Finds the minimum value of a data set.
`static double`	`computePercentile(java.util.Collection<? extends java.lang.Number> data, double percentile)` Computes the percentile value of the given data.
`static double[]`	`computePercentiles(java.util.Collection<? extends java.lang.Number> data, double... percentiles)` Computes the given percentiles of the given data.
`static double`	`computeRootMeanSquaredError(java.util.Collection<? extends java.lang.Number> data)` Computes the Root mean-squared (RMS) error between the data and its mean.
`static double`	`computeRootMeanSquaredError(java.util.Collection<? extends java.lang.Number> data, double mean)` Computes the Root mean-squared (RMS) error between the data and its mean
`static double`	`computeSkewness(java.util.Collection<? extends java.lang.Number> data)` Computes the unbiased skewness of the dataset.
`static double`	`computeStandardDeviation(java.util.Collection<? extends java.lang.Number> data)` Computes the standard deviation of a dataset, which is the square root of the unbiased variance.
`static double`	`computeStandardDeviation(java.util.Collection<? extends java.lang.Number> data, double mean)` Computes the standard deviation of a dataset, which is the square root of the unbiased variance.
`static double`	`computeSum(java.lang.Iterable<? extends java.lang.Number> data)` Computes the arithmetic sum of the dataset
`static double`	`computeSumSquaredDifference(java.lang.Iterable<? extends java.lang.Number> data, double target)` Computes the sum-squared difference between the data and a target
`static double`	`computeVariance(java.util.Collection<? extends java.lang.Number> data)` Computes the unbiased variance (second central moment, squared standard deviation) of a dataset.
`static double`	`computeVariance(java.util.Collection<? extends java.lang.Number> data, double mean)` Computes the unbiased variance (second central moment, squared standard deviation) of a dataset
`static double`	`computeWeightedCentralMoment(java.lang.Iterable<? extends WeightedValue<? extends java.lang.Number>> data, double mean, int moment)` Computes the desired biased estimate central moment of the given dataset.
`static double`	`computeWeightedKurtosis(java.util.Collection<? extends WeightedValue<? extends java.lang.Number>> data)` Computes the biased excess kurtosis of the given dataset.
`static double`	`computeWeightedMean(java.lang.Iterable<? extends WeightedValue<? extends java.lang.Number>> data)` Computes the arithmetic mean (average, expectation, first central moment) of a dataset.
`static Pair<java.lang.Double,java.lang.Double>`	`computeWeightedMeanAndVariance(java.lang.Iterable<? extends WeightedValue<? extends java.lang.Number>> data)` Computes the mean and unbiased variance of a Collection of data using the one-pass approach.
`static double`	`getPercentileFromSorted(java.util.List<? extends java.lang.Number> sortedData, double percentile)` Computes the percentile value of the given pre-sorted data in increasing order.

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Constructor Detail
  - UnivariateStatisticsUtil
```
public UnivariateStatisticsUtil()
```
- Method Detail
  - computeMean
```
@PublicationReference(title="Algorithms for calculating variance",
                      type=WebPage,
                      year=2010,
                      author="Wikipedia",
                      url="http://en.wikipedia.org/wiki/Algorithms_for_calculating_variance")
public static double computeMean(java.lang.Iterable<? extends java.lang.Number> data)
```
    Computes the arithmetic mean (average, expectation, first central moment) of a dataset
    
    Parameters:
    
    data - Collection of Doubles to consider
    
    Returns:
    
    Arithmetic mean of the given dataset
  - computeWeightedMean
```
@PublicationReference(title="Algorithms for calculating variance",
                      type=WebPage,
                      year=2010,
                      author="Wikipedia",
                      url="http://en.wikipedia.org/wiki/Algorithms_for_calculating_variance")
public static double computeWeightedMean(java.lang.Iterable<? extends WeightedValue<? extends java.lang.Number>> data)
```
    Computes the arithmetic mean (average, expectation, first central moment) of a dataset. The absolute value of the weight is used to handle negative weights.
    
    Parameters:
    
    data - Collection of Doubles to consider.
    
    Returns:
    
    Arithmetic mean of the given dataset.
  - computeVariance
```
public static double computeVariance(java.util.Collection<? extends java.lang.Number> data)
```
    Computes the unbiased variance (second central moment, squared standard deviation) of a dataset. Computes the mean first, then computes the variance. If you already have the mean, then use the two-argument computeVariance(data,mean) method to save duplication of effort.
    
    Parameters:
    
    data - Data to consider
    
    Returns:
    
    Unbiased variance of the given dataset
  - computeVariance
```
public static double computeVariance(java.util.Collection<? extends java.lang.Number> data,
                                     double mean)
```
    Computes the unbiased variance (second central moment, squared standard deviation) of a dataset
    
    Parameters:
    
    data - Data to consider
    
    mean - Pre-computed mean (or central value) of the dataset
    
    Returns:
    
    Unbiased variance of the given dataset
  - computeStandardDeviation
```
public static double computeStandardDeviation(java.util.Collection<? extends java.lang.Number> data)
```
    Computes the standard deviation of a dataset, which is the square root of the unbiased variance. It computes the mean first and then computes the standard deviation. If you already have the mean, then use the two-argument computeStandardDeviation(data, mean) to save duplication of effort.
    
    Parameters:
    
    data - The data to consider.
    
    Returns:
    
    The standard deviation of the given data. Will not be negative.
  - computeStandardDeviation
```
public static double computeStandardDeviation(java.util.Collection<? extends java.lang.Number> data,
                                              double mean)
```
    Computes the standard deviation of a dataset, which is the square root of the unbiased variance.
    
    Parameters:
    
    data - The data to consider.
    
    mean - The pre-computed mean of the given data.
    
    Returns:
    
    The standard deviation of the given data. Will not be negative.
  - computeRootMeanSquaredError
```
public static double computeRootMeanSquaredError(java.util.Collection<? extends java.lang.Number> data)
```
    Computes the Root mean-squared (RMS) error between the data and its mean. Computes the mean first, then computes the RMS error. If you already have the mean, then use the two-argument computeRootMeanSquaredError(data,mean) method to save computation
    
    Parameters:
    
    data - Dataset to consider
    
    Returns:
    
    RMS error of the dataset about the mean
  - computeRootMeanSquaredError
```
public static double computeRootMeanSquaredError(java.util.Collection<? extends java.lang.Number> data,
                                                 double mean)
```
    Computes the Root mean-squared (RMS) error between the data and its mean
    
    Parameters:
    
    data - Dataset to consider
    
    mean - Mean value about which to compute the sum-squared error
    
    Returns:
    
    RMS error of the dataset about the mean
  - computeSum
```
public static double computeSum(java.lang.Iterable<? extends java.lang.Number> data)
```
    Computes the arithmetic sum of the dataset
    
    Parameters:
    
    data - Dataset to consider
    
    Returns:
    
    Arithmetic sum of the given dataset
  - computeSumSquaredDifference
```
public static double computeSumSquaredDifference(java.lang.Iterable<? extends java.lang.Number> data,
                                                 double target)
```
    Computes the sum-squared difference between the data and a target
    
    Parameters:
    
    data - Dataset to consider
    
    target - Target about which to compute the difference
    
    Returns:
    
    Sum-squared difference between the dataset and the target
  - computeCorrelation
```
@PublicationReference(author="Wikipedia",
                      title="Pearson product-moment correlation coefficient",
                      type=WebPage,
                      year=2011,
                      url="http://en.wikipedia.org/wiki/Pearson_product-moment_correlation_coefficient")
public static double computeCorrelation(java.util.Collection<? extends java.lang.Number> data1,
                                                                                                                                                                                                                                                                                                                                              java.util.Collection<? extends java.lang.Number> data2)
```
    Computes the correlation coefficient in a single pass. However, this algorithm can become numerically unstable but is about twice as fast as the non-single-pass method.
    
    Parameters:
    
    data1 - First dataset to consider, must have same size as data2
    
    data2 - Second dataset to consider
    
    Returns:
    
    Normalized correlation coefficient, [-1,+1]
  - computeMedian
```
public static double computeMedian(java.util.Collection<? extends java.lang.Number> data)
```
    Computes the median of the given data.
    
    Parameters:
    
    data - Data from which to compute the median.
    
    Returns:
    
    Median of the sample.
  - computePercentile
```
public static double computePercentile(java.util.Collection<? extends java.lang.Number> data,
                                       double percentile)
```
    Computes the percentile value of the given data. For example, if data has 101 values and "percentile" is 0.27, then the return value would be the ascending-sorted value in the 26th zero-based index.
    
    Parameters:
    
    data - Data from which to compute the percentile.
    
    percentile - Percentile to choose, must be on the closed interval 0.0 to 1.0.
    
    Returns:
    
    Requested percentile from the data.
  - getPercentileFromSorted
```
public static double getPercentileFromSorted(java.util.List<? extends java.lang.Number> sortedData,
                                             double percentile)
```
    Computes the percentile value of the given pre-sorted data in increasing order. For example, if data has 101 values and "percentile" is 0.27, then the return value would be the ascending-sorted value in the 26th zero-based index.
    
    Parameters:
    
    sortedData - The data sorted in increasing order from which to get the percentile. Must have at least one element.
    
    percentile - The percentile to choose. Must be between 0 and 1, inclusive.
    
    Returns:
    
    The requested percentile from the data.
  - computePercentiles
```
public static double[] computePercentiles(java.util.Collection<? extends java.lang.Number> data,
                                          double... percentiles)
```
    Computes the given percentiles of the given data. For example, if data has 101 values and "percentile" is 0.27, then the return value would be the ascending-sorted value in the 26th zero-based index.
    
    Parameters:
    
    data - The data sorted from which to compute the percentiles. Must have at least one element.
    
    percentiles - The percentiles to compute. All must be between 0 and 1, inclusive.
    
    Returns:
    
    The an array of equal length to the given percentiles that contains the requested percentiles from the data.
  - computeMinimum
```
public static double computeMinimum(java.lang.Iterable<? extends java.lang.Number> data)
```
    Finds the minimum value of a data set.
    
    Parameters:
    
    data - Data set to consider
    
    Returns:
    
    Minimum value of the data set.
  - computeMaximum
```
public static double computeMaximum(java.lang.Iterable<? extends java.lang.Number> data)
```
    Finds the maximum value of a data set.
    
    Parameters:
    
    data - Data set to consider
    
    Returns:
    
    Maximum value of the data set.
  - computeMinAndMax
```
public static Pair<java.lang.Double,java.lang.Double> computeMinAndMax(java.lang.Iterable<? extends java.lang.Number> data)
```
    Computes the minimum and maximum of a set of data in a single pass.
    
    Parameters:
    
    data - Data to consider
    
    Returns:
    
    Minimum and Maximum
  - computeSkewness
```
@PublicationReference(author="Wikipedia",
                      title="Skewness",
                      type=WebPage,
                      year=2009,
                      url="http://en.wikipedia.org/wiki/Skewness")
public static double computeSkewness(java.util.Collection<? extends java.lang.Number> data)
```
    Computes the unbiased skewness of the dataset.
    
    Parameters:
    
    data - Data from which to compute the unbiased skewness.
    
    Returns:
    
    Unbiased skewness.
  - computeCentralMoment
```
public static double computeCentralMoment(java.lang.Iterable<? extends java.lang.Number> data,
                                          double mean,
                                          int moment)
```
    Computes the desired biased estimate central moment of the given dataset.
    
    Parameters:
    
    data - Data to compute the moment of.
    
    mean - Mean of the data (to prevent redundant computation).
    
    moment - Desired moment of the data, must be greater than or equal to 1.
    
    Returns:
    
    Biased estimate of the desired central moment.
  - computeWeightedCentralMoment
```
public static double computeWeightedCentralMoment(java.lang.Iterable<? extends WeightedValue<? extends java.lang.Number>> data,
                                                  double mean,
                                                  int moment)
```
    Computes the desired biased estimate central moment of the given dataset. The absolute value of the weight is used to handle negative weights.
    
    Parameters:
    
    data - Data to compute the moment of.
    
    mean - Mean of the data (to prevent redundant computation).
    
    moment - Desired moment of the data, must be greater than or equal to 1.
    
    Returns:
    
    Biased estimate of the desired central moment.
  - computeKurtosis
```
@PublicationReference(author="Wikipedia",
                      title="Kurtosis",
                      type=WebPage,
                      year=2009,
                      url="http://en.wikipedia.org/wiki/Kurtosis")
public static double computeKurtosis(java.util.Collection<? extends java.lang.Number> data)
```
    Computes the biased excess kurtosis of the given dataset. Intuitively, kurtosis quantifies the pointiness of the data by normalizing the fourth central moment.
    
    Parameters:
    
    data - Dataset to compute its kurtosis.
    
    Returns:
    
    Biased excess kurtosis of the given dataset.
  - computeWeightedKurtosis
```
@PublicationReference(author="Wikipedia",
                      title="Kurtosis",
                      type=WebPage,
                      year=2009,
                      url="http://en.wikipedia.org/wiki/Kurtosis")
public static double computeWeightedKurtosis(java.util.Collection<? extends WeightedValue<? extends java.lang.Number>> data)
```
    Computes the biased excess kurtosis of the given dataset. Intuitively, kurtosis quantifies the pointiness of the data by normalizing the fourth central moment. The absolute value of the weight is used to handle negative weights.
    
    Parameters:
    
    data - Dataset to compute its kurtosis.
    
    Returns:
    
    Biased excess kurtosis of the given dataset.
  - computeEntropy
```
@PublicationReference(author="Wikipedia",
                      title="Entropy (information theory)",
                      type=WebPage,
                      year=2009,
                      url="http://en.wikipedia.org/wiki/Entropy_(Information_theory)")
public static double computeEntropy(java.lang.Iterable<? extends java.lang.Number> data)
```
    Computes the information-theoretic entropy of the PMF in bits (base 2).
    
    Parameters:
    
    data - Data to compute the entropy.
    
    Returns:
    
    Entropy in bits of the given PMF.
  - computeMeanAndVariance
```
@PublicationReference(title="Algorithms for calculating variance",
                      type=WebPage,
                      year=2010,
                      author="Wikipedia",
                      url="http://en.wikipedia.org/wiki/Algorithms_for_calculating_variance")
public static Pair<java.lang.Double,java.lang.Double> computeMeanAndVariance(java.lang.Iterable<? extends java.lang.Number> data)
```
    Computes the mean and unbiased variance of a Collection of data using the one-pass approach.
    
    Parameters:
    
    data - Data to consider
    
    Returns:
    
    Mean and unbiased Variance Pair.
  - computeWeightedMeanAndVariance
```
@PublicationReference(title="Algorithms for calculating variance",
                      type=WebPage,
                      year=2010,
                      author="Wikipedia",
                      url="http://en.wikipedia.org/wiki/Algorithms_for_calculating_variance")
public static Pair<java.lang.Double,java.lang.Double> computeWeightedMeanAndVariance(java.lang.Iterable<? extends WeightedValue<? extends java.lang.Number>> data)
```
    Computes the mean and unbiased variance of a Collection of data using the one-pass approach. The absolute value is used to handle negative weights.
    
    Parameters:
    
    data - Data to consider.
    
    Returns:
    
    Mean and unbiased Variance Pair.

Class UnivariateStatisticsUtil

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Constructor Detail

UnivariateStatisticsUtil

Method Detail

computeMean

computeWeightedMean

computeVariance

computeVariance

computeStandardDeviation

computeStandardDeviation

computeRootMeanSquaredError

computeRootMeanSquaredError

computeSum

computeSumSquaredDifference

computeCorrelation

computeMedian

computePercentile

getPercentileFromSorted

computePercentiles

computeMinimum

computeMaximum

computeMinAndMax

computeSkewness

computeCentralMoment

computeWeightedCentralMoment

computeKurtosis

computeWeightedKurtosis

computeEntropy

computeMeanAndVariance

computeWeightedMeanAndVariance