v.24.1New Feature

Added seriesDecomposeSTL function for time series decomposition

Official CH changelog text
Added function seriesDecomposeSTL which decomposes a time series into a season, a trend and a residual component. #57078 (Bhavna Jindal).
Introduced the seriesDecomposeSTL function to decompose a time series into seasonal, trend, and residual components using STL (Seasonal-Trend decomposition using Loess).

Why it matters

This feature enables users to analyze and understand the underlying patterns within time series data by separating it into meaningful parts: seasonality, trend, and noise. It helps in better forecasting, anomaly detection, and time series analysis directly within ClickHouse.

How to use it

Apply the seriesDecomposeSTL function to a time series dataset by passing the required parameters. The function returns the decomposed components that you can use for further analysis or visualization.

Related resources

PreviousNext