release_lag_rule#

Back to L1 | Browse all axes | Browse all options

Axis release_lag_rule on sub-layer l1_c (layer l1).

Sub-layer#

l1_c

Axis metadata#

  • Default: 'ignore_release_lag'

  • Sweepable: False

  • Status: operational

Operational status summary#

  • Operational: 3 option(s)

  • Future: 0 option(s)

Options#

ignore_release_lag – operational#

Treat every observation as available at its calendar period.

Pseudo-real-time mode: ignores the release-lag distinction; every variable is assumed to be available the moment the period closes.

When to use

Backtests where real-time vintage data is unavailable.

References

  • macroforecast design Part 1, L1: ‘data definition is the recipe layer that pins source, target, geography, and horizon – everything downstream branches off these choices.’

Related options: fixed_lag_all_series, series_specific_lag

Last reviewed 2026-05-05 by macroforecast author.

fixed_lag_all_series – operational#

Apply a single release lag to every series.

All series shift by leaf_config.fixed_lag_periods periods. Approximates real-time availability without per-series detail.

When to use

Coarse real-time approximations.

References

  • macroforecast design Part 1, L1: ‘data definition is the recipe layer that pins source, target, geography, and horizon – everything downstream branches off these choices.’

Related options: series_specific_lag

Parameters

name

type

default

constraint

description

fixed_lag_periods

int

None

>=0; optional; defaults to 0 if not set.

Uniform release lag in periods applied to every predictor series. A value of 1 means each series is available one period after the period it was observed.

Last reviewed 2026-05-05 by macroforecast author.

series_specific_lag – operational#

Use per-series release lags from leaf_config.

Honours the published release-lag table in leaf_config.release_lag_per_series. Most accurate option for true real-time studies.

When to use

Real-time / nowcasting studies that respect publication delays.

References

  • macroforecast design Part 1, L1: ‘data definition is the recipe layer that pins source, target, geography, and horizon – everything downstream branches off these choices.’

Related options: fixed_lag_all_series

Parameters

name

type

default

constraint

description

release_lag_per_series

dict[str, int]

—

Required when release_lag_rule=series_specific_lag; non-empty dict.

Per-series release lag in periods. Maps series name to a non-negative integer. Series not present in the dict are treated as zero-lag (available immediately).

Last reviewed 2026-05-05 by macroforecast author.