Lag Selection for Univariate Time Series Forecasting using Deep Learning: An Empirical Study
This addresses lag selection for time series forecasting using deep learning, but it is incremental as it extends existing methods to new contexts without introducing novel techniques.
The study tackled the problem of selecting the number of past observations (lags) for univariate time series forecasting with deep learning, finding that both too small and too large lag sizes negatively impact performance, with cross-validation performing best but comparable to simple heuristics.
Most forecasting methods use recent past observations (lags) to model the future values of univariate time series. Selecting an adequate number of lags is important for training accurate forecasting models. Several approaches and heuristics have been devised to solve this task. However, there is no consensus about what the best approach is. Besides, lag selection procedures have been developed based on local models and classical forecasting techniques such as ARIMA. We bridge this gap in the literature by carrying out an extensive empirical analysis of different lag selection methods. We focus on deep learning methods trained in a global approach, i.e., on datasets comprising multiple univariate time series. The experiments were carried out using three benchmark databases that contain a total of 2411 univariate time series. The results indicate that the lag size is a relevant parameter for accurate forecasts. In particular, excessively small or excessively large lag sizes have a considerable negative impact on forecasting performance. Cross-validation approaches show the best performance for lag selection, but this performance is comparable with simple heuristics.