# solarforecastarbiter.reference_forecasts.persistence¶

Functions for persistence forecasts.

Two kinds of persistence are supported:

1. Persistence of observed values in persistence_scalar() and persistence_interval()
2. Persistence of irradiance or power accounting for solar position in persistence_scalar_index() and persistence_interval_index() (?).

Users of intraday persistence forecasts will typically want to use persistence_scalar() or persistence_scalar_index(). Users of day ahead persistence forecasts will typically want to use persistence_interval(). persistence_interval_index()?

The functions accept a load_data keyword argument that allows users to change where the functions load the observation data from. This is most useful for users that would like to provide their own observation data rather than using the solarforecastarbiter database.

Functions

 persistence_interval(observation, …) Make a persistence forecast for an observation using the mean values of each interval_length bin from data_start to data_end. persistence_probabilistic(observation, …) Make a probabilistic persistence forecast using the observation from data_start to data_end. persistence_probabilistic_timeofday(…) Make a probabilistic persistence forecast using the observation from data_start to data_end, matched by time of day (e.g. persistence_scalar(observation, data_start, …) Make a persistence forecast using the mean value of the observation from data_start to data_end. persistence_scalar_index(observation, …) Calculate a persistence forecast using the mean value of the observation clear sky index or AC power index from data_start to data_end.