rustkernel-temporal 0.1.1

RustKernels Temporal domain kernels
Documentation

rustkernel-temporal

Crates.io Documentation License

GPU-accelerated temporal analysis kernels for forecasting, decomposition, detection, and volatility modeling.

Kernels (7)

Forecasting (2 kernels)

  • ARIMAForecast - ARIMA(p,d,q) model fitting and forecasting
  • ProphetDecomposition - Prophet-style trend/seasonal/holiday decomposition

Detection (2 kernels)

  • ChangePointDetection - PELT/Binary segmentation/CUSUM
  • TimeSeriesAnomalyDetection - Statistical threshold detection

Volatility (2 kernels)

  • GARCHVolatility - GARCH(1,1) volatility modeling
  • RealizedVolatility - High-frequency realized volatility

Seasonality (1 kernel)

  • SeasonalDecomposition - STL decomposition

Installation

Add to your Cargo.toml:

[dependencies]
rustkernel-temporal = "0.1.0"

Usage

use rustkernel_temporal::prelude::*;

// Forecast with ARIMA
let arima = ARIMAForecast::new();
let forecast = arima.fit_and_forecast(&series, p, d, q, horizon);

License

Apache-2.0