numeraire.core.simulate.RebalanceSchedule#

class numeraire.core.simulate.RebalanceSchedule(data_calendar: DatetimeIndex, signal_dates: DatetimeIndex, spans: tuple[tuple[int, int], ...])[source]#

Bases: object

Decision (signal) dates mapped to the half-open data-row spans they govern.

spans[k] = (lo, hi) means the k-th signal’s target holds over data rows lo..hi-1 — lo is the first return row strictly after the signal date, hi is the next signal’s first row (or the end of data). Decouples the decision calendar from the data frequency (e.g. month-end decisions over daily returns).

__init__(data_calendar: DatetimeIndex, signal_dates: DatetimeIndex, spans: tuple[tuple[int, int], ...]) → None#

Methods

`__init__`(data_calendar, signal_dates, spans)
`from_rule`(data_calendar[, rule])	Derive signal dates from the data calendar (`month_end`: last data date per month).
`from_signals`(data_calendar, signal_dates)	Schedule from explicit signal dates (each trades on the next data row after it).

Attributes

`data_calendar`
`signal_dates`
`spans`

classmethod from_signals(data_calendar: DatetimeIndex, signal_dates: DatetimeIndex) → RebalanceSchedule[source]#: Schedule from explicit signal dates (each trades on the next data row after it).

classmethod from_rule(data_calendar: DatetimeIndex, rule: str = 'month_end') → RebalanceSchedule[source]#: Derive signal dates from the data calendar (month_end: last data date per month).