Features:

• From date: day, month, year, weekday, week_num, is_holiday
• Lag metric: same metric from 1 day ago, or 7 days ago, etc.
• Moving average
• Lags on subsets of the specified group columns (e.g. {Store, Department} vs. {Department} vs. {Store})
• Exponentially Weighted Moving Averages (EWMA) of n-th order differentiated lags
• Aggregation of lags (mean, std, sums, etc.)
• Interactions of lags (e.g. Lag2 - Lag1)
• Linear regression on lags (taking slope and/or intercept as new features)

Choosing the lag interval:

• Ranking based on autocorrelation
• Pre-defined intervals (based on estimated frequency; weeks, days, etc.)

See also:

• Ways to prevent overfitting in time series prediction

References

https://www.slideshare.net/0xdata/marios-michailidis-mathias-muller-h2oai-time-series-with-h2o-driverless-ai-h2o-world-san-francisco

https://youtu.be/0pvvDHfxdZ8