fc_binned: A binned probability forecast

In kdayday/forecasting: Probabilistic Solar Forecsting

Description

Initialize a univariate probabilistic power forecast for a specific time point by ranking ensemble members and *linearly interpolating*. For a classic stepped forecast, see fc_empirical.Its rank quantiles are the basic quantiles estimated from the ensemble members; the quantiles are the rank quantiles iterpolated to the quantiles of interest (i.e., 10 etc...).

Usage

fc_binned(data.input, location, time, max_power, ...)

Arguments

data.input	A numeric vector of ensemble members
location	A string
time	A lubridate time stamp
max_power	Maximum power for normalizing forecast to [0,1]

Value

A probabilistic forecast object

See Also

Other forecast classes: fc_bma, fc_emos, fc_empirical

kdayday/forecasting documentation built on Oct. 7, 2020, 7:16 p.m.