We integrate best practices in time series modeling, including automatic time series feature engineering to discover predicted signals. It can also automatically detect smoothness, seasonality, switch targets and perform backtests for maximum accuracy.
In addition to basic and proven time series methods like ARIMA and Facebook Prophet, we also include advanced time series models to help you achieve higher prediction accuracy.
Because the goal of time series models is to extract understanding and predict future results, we provide multiple ways to visualize insights over time and deploy models into production, including integrating models into business processes and applications. Full API support.