lags_adder

Lag feature generation for time series forecasting.

Creates lagged versions of target variables to capture temporal patterns. Supports multiple strategies: trivial lags (minute/day-based), custom lags, and autocorrelation-based lags for adaptive feature engineering.

Functions

generate_autocorr_lags(signal, min_horizon)	Generate lag features based on autocorrelation peaks in the time series.
generate_day_lags(min_horizon, max_day_lags)	Generate day-based lag features for capturing daily and weekly patterns.
generate_minute_lags(min_horizon)	Generate minute-based lag features for short-term forecasting.

Classes

LagsAdder(**data)

Transform that adds lag features to time series data.