Skip to content

n_lags Lagged regressor to n_forecasts forecast steps ratio #320

Discussion options

You must be logged in to vote

@veioxve3r1d23seb2g6b

What does n_lag really mean? Lagging by n rows?

n_lags is the number of most recent observations to include in the model. e.g. n_lag=3 uses all of the last 3 hours as input to the model.

Should n_lags always be greater than n_forecasts?

There is no rule set in stone. Generally, it should be at least same.

What is the optimal ratio when using n_forecasts and n_lags?

The conventional ratio is around 1:1 to 1:2, sometimes higher.
e.g. hourly data: when predicting next 6h, use last 24h
e.g. hourly data: when predicting next 72h, use last 72h

Replies: 1 comment

Comment options

You must be logged in to vote
0 replies
Answer selected by ourownstory
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
1 participant