logscore: Calculate the log predictive score

In lily940703/febama: Feature-based Bayesian Forecasting Model Averaging

Calculate the log predictive score

Description

Calculate the log predictive score of a time series for FEBAMA framework. The weights in forecast combination are related to time series features.

Usage

logscore(data, beta, betaIdx, features_used, sum = TRUE)

Arguments

data	A list with lpd and feat (the output of function lpd_features_multi).
beta	A list of coefficient vectors of the features.
betaIdx	A list of indicator vectors.
features_used	The features used for forecast combination. See the parameter seting for more information.
sum	If TRUE, return the sume of log predictive densities.

Value

logscore returns the value of log predictive score.

lily940703/febama documentation built on March 20, 2022, 1:57 a.m.