Class TimeSeries

java.lang.Object

org.episteme.core.mathematics.statistics.timeseries.TimeSeries

public class TimeSeries
extends Object

Time series analysis and forecasting.

ARIMA: AutoRegressive Integrated Moving Average Components: AR(p), I(d), MA(q)

Since:

1.0

Author:

Silvere Martin-Michiellot, Gemini AI (Google DeepMind)

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	TimeSeries.ARModel	AutoRegressive model AR(p): X_t = c + Σφ_i*X_(t-i) + ε_t
static class	TimeSeries.MAModel	Moving Average model MA(q): X_t = μ + ε_t + Σθ_i*ε_(t-i)
static class	TimeSeries.SeasonalDecomposition	Seasonal decomposition: Trend + Seasonal + Residual.

Constructor Summary

Constructors

Constructor	Description
TimeSeries()

Method Summary

Modifier and Type	Method	Description
static Real	autocorrelation(Real[] data, int lag)	Computes autocorrelation at lag k.
static Real[]	difference(Real[] data, int order)	Differencing to achieve stationarity.
static Real[]	exponentialMovingAverage(Real[] data, Real alpha)	Exponential moving average (EMA).
static Real[]	movingAverage(Real[] data, int window)	Simple moving average (SMA).

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

TimeSeries

public TimeSeries()

Method Details

movingAverage

public static Real[] movingAverage(Real[] data,
 int window)

Simple moving average (SMA).

Parameters:

data - time series data

window - window size

Returns:

smoothed series

exponentialMovingAverage

public static Real[] exponentialMovingAverage(Real[] data,
 Real alpha)

Exponential moving average (EMA).

EMA_t = α * X_t + (1-α) * EMA_(t-1)

Parameters:

data - time series

alpha - smoothing factor (0 invalid input: '<' α invalid input: '<' 1)

Returns:

exponentially smoothed series

autocorrelation

public static Real autocorrelation(Real[] data,
 int lag)

Computes autocorrelation at lag k.

ρ_k = Cov(X_t, X_(t-k)) / Var(X_t)

difference

public static Real[] difference(Real[] data,
 int order)

Differencing to achieve stationarity.

Δ^d X_t = X_t - X_(t-1) applied d times