search

petersonR / fastTS

Fast, effective, time series modeling with seasonality and exogenous variables via the sparsity-ranked lasso
https://petersonr.github.io/fastTS/
GNU General Public License v3.0
3 stars 0 forks source link

Improvement: more ways of adding endogenous non-autoregressive predictors #3

Open petersonR opened 7 months ago

petersonR commented 7 months ago

Some series might benefit from additional global endogenous terms, such as time-based indicators (year/month/weekday/holiday) or trend/spline functionality. These are "predictors" that are known at all possible times in past + future, and could be included as additional terms in the model matrix by default (or with modifiable options).

petersonR commented 7 months ago

This can currently be done "by hand" but is clunky.