movingPercentileBy()
BigDecimal objects mapped from a Stream<INPUT> via a mappingFunction, looking back windowSize elements.Implementation Notes
This implementation is suitable for mapping an arbitrary Stream<INPUT> to BigDecimal via a mappingFunction; for a version that operates directly on a Stream<BigDecimal>, see movingPercentile().
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).
Values that fall between two ranks are linearly interpolated (the R-7 method, NumPy’s default). The percentile is converted to BigDecimal from its decimal representation, so 99.9 is exactly 99.9. The rank and interpolation are calculated exactly, and only the result is rounded to the MathContext.
Signatures
movingPercentileBy(double percentile, int windowSize, Function<INPUT, BigDecimal> mappingFunction)
percentile- The percentile to calculate, between 0 and 100, inclusivewindowSize- How many trailing elements to calculate the percentile over, must be greater than 1mappingFunction- A non-null function to map streamINPUTelements intoBigDecimalfor calculation
Additional Methods
| Method | Purpose |
|---|---|
excludePartialValues() | When calculating the moving percentile, and the full size of the window has not yet been reached, the gatherer should suppress emitting values until the lookback window is full. |
treatNullAsZero() | When encountering a null value in a stream, treat it as BigDecimal.ZERO instead. |
treatNullAs(BigDecimal replacement) | When encountering a null value in a stream, treat it as the given replacement value instead. |
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. |
withOriginal() | Include the original input value from the stream in addition to the calculated value in a WithOriginal record. See example. |
See movingPercentile() for examples of each of these.
Examples
Moving 90th percentile of window size 3, mapped from an object
record NamedValue(String name, BigDecimal value) {}
Stream
.of(
new NamedValue("first", new BigDecimal("1.0")),
new NamedValue("second", new BigDecimal("2.0")),
new NamedValue("third", new BigDecimal("10.0")),
new NamedValue("fourth", new BigDecimal("2.0"))
)
.gather(Gatherers4j.movingPercentileBy(90, 3, NamedValue::value))
.toList();
// [
// BigDecimal("1.0"),
// BigDecimal("1.9"),
// BigDecimal("8.4"),
// BigDecimal("8.4")
// ]
Emitting a record containing the original and calculated values
record NamedValue(String name, BigDecimal value) {}
Stream
.of(
new NamedValue("first", new BigDecimal("1.0")),
new NamedValue("second", new BigDecimal("2.0")),
new NamedValue("third", new BigDecimal("10.0"))
)
.gather(Gatherers4j.movingPercentileBy(90, 2, NamedValue::value).withOriginal())
.toList();
// [
// WithOriginal[original=NamedValue[name=first, value=1.0], calculated=1.0],
// WithOriginal[original=NamedValue[name=second, value=2.0], calculated=1.9],
// WithOriginal[original=NamedValue[name=third, value=10.0], calculated=9.2]
// ]