Expand your Time Series Arsenal with These Models

Regularizing, Bagging, Stacking, and More

Time series data typically has four components:

• Autoregression
• Seasonality
• Trend
• Residual

Predict these components, and you can forecast almost any time series. Sounds easy enough, right?

Well, not exactly. There is a lot of ambiguity around the best ways to specify a model so that it can account for these elements correctly, and much research in the past years has been released to find the best ways to do so, with state-of-the …

easy ensemble-learning forecasting machine learning series time time series time-series-analysis