movingVariance()

Calculate the moving variance of a Stream<BigDecimal> looking back windowSize elements, as either a population or a sample.

Implementation Notes

This implementation is suitable for Stream<BigDecimal>, for a version that takes a user-specified mapping function see movingVarianceBy(). By default, nulls are ignored and play no part in calculations, see treatNullAs() and treatNullAsZero() below for ways to change this behavior. The default MathContext for all calculations is MathContext.DECIMAL64, but this can be overridden (see withMathContext(), below).

Use Dataset.Population when the elements are the entire dataset (the variance is divided by n), and Dataset.Sample when they are a sample of a larger population (the variance is divided by n - 1). The sample variance of a single value is undefined, so when using Dataset.Sample nothing is emitted until at least two values have been seen.

See also movingStandardDeviation(), which is the square root of this value.

Signatures

movingVariance(Dataset dataset, int windowSize)

dataset - Either Population or Sample
windowSize - How many trailing elements to calculate the variance over, must be greater than 1

Additional Methods

Method	Purpose
`excludePartialValues()`	When calculating the moving variance, and the full size of the window has not yet been reached, the gatherer should suppress emitting values until the lookback window is full. See example.
`treatNullAsZero()`	When encountering a `null` value in a stream, treat it as `BigDecimal.ZERO` instead. See example.
`treatNullAs(BigDecimal replacement)`	When encountering a `null` value in a stream, treat it as the given `replacement` value instead. See example.
`treatNullAsOne()`	When encountering a `null` value in a stream, treat it as `BigDecimal.ONE` instead.
`withMathContext(MathContext mathContext)`	Replace the `MathContext` used for all mathematical operations performed by this gatherer. See example.
`withOriginal()`	Include the original input value from the stream in addition to the calculated value in a `WithOriginal` record. See example.

Examples

Moving variance (population) of window size 3

Stream
    .of("1.0", "2.0", "10.0", "2.0")
    .map(BigDecimal::new)
    .gather(Gatherers4j.movingVariance(Dataset.Population, 3))
    .toList();

// [
//   BigDecimal("0.00")
//   BigDecimal("0.25")
//   BigDecimal("16.22222222222222")
//   BigDecimal("14.22222222222222")
// ]

Moving variance (sample) of window size 3

Stream
    .of("1.0", "2.0", "10.0", "2.0")
    .map(BigDecimal::new)
    .gather(Gatherers4j.movingVariance(Dataset.Sample, 3))
    .toList();

// [
//   BigDecimal("0.50")
//   BigDecimal("24.33333333333333")
//   BigDecimal("21.33333333333333")
// ]

Excluding partial values

Showing that in-process moving variance values are not emitted for each element until the lookback window has been filled.

Stream
    .of("1.0", "2.0", "10.0", "2.0")
    .map(BigDecimal::new)
    .gather(Gatherers4j.movingVariance(Dataset.Population, 3).excludePartialValues())
    .toList();

// [
//   BigDecimal("16.22222222222222")
//   BigDecimal("14.22222222222222")
// ]

Showing nulls are ignored by default

Stream
    .of(null, null, new BigDecimal("10.0"), new BigDecimal("2.0"), new BigDecimal("1.0"))
    .gather(Gatherers4j.movingVariance(Dataset.Population, 3))
    .toList();

// [
//   BigDecimal("0.00")
//   BigDecimal("16.00")
//   BigDecimal("16.22222222222222")
// ]

Treating null as zero

Stream
    .of(null, null, new BigDecimal("10.0"), new BigDecimal("2.0"))
    .gather(Gatherers4j.movingVariance(Dataset.Population, 3).treatNullAsZero())
    .toList();

// [
//   BigDecimal("0")
//   BigDecimal("0")
//   BigDecimal("22.22222222222222")
//   BigDecimal("18.66666666666667")
// ]

Replacing null with another `BigDecimal`

Stream
    .of(null, null, new BigDecimal("10.0"), new BigDecimal("2.0"))
    .gather(Gatherers4j.movingVariance(Dataset.Population, 3).treatNullAs(new BigDecimal("3.5")))
    .toList();

// [
//   BigDecimal("0.00")
//   BigDecimal("0.00")
//   BigDecimal("9.388888888888889")
//   BigDecimal("12.05555555555556")
// ]

Specifying a new `MathContext`

Stream
    .of("1.0", "2.0", "10.0", "2.0")
    .map(BigDecimal::new)
    .gather(Gatherers4j
        .movingVariance(Dataset.Population, 3)
        .withMathContext(new MathContext(3, RoundingMode.DOWN))
    )
    .toList();

// [
//   BigDecimal("0.00")
//   BigDecimal("0.25")
//   BigDecimal("16.2")
//   BigDecimal("14.2")
// ]

Emitting a record containing the original and calculated values

Stream
    .of("1.0", "2.0", "10.0", "2.0")
    .map(BigDecimal::new)
    .gather(Gatherers4j.movingVariance(Dataset.Population, 3).withOriginal())
    .toList();

// [
//   WithOriginal[original=1.0, calculated=0.00]
//   WithOriginal[original=2.0, calculated=0.25]
//   WithOriginal[original=10.0, calculated=16.22222222222222]
//   WithOriginal[original=2.0, calculated=14.22222222222222]
// ]