caliper_relevance_new: Caliper method for local weights, second iteration
In ooelrich/oscbvar: Stuff for Bayesian vector autoregressions

Calculates weights for which previous log scores to use based on the caliper method. This is an implementation of the simpel caliper method where local predictive ability is the sum of relevant log scores. In case there are no relevant data to base the measure on, each observation gets weight zero, and so each model will have equal weight.

1	caliper_relevance_new(atomic_df, sotw, start_agg = 161, cw = 5, matching_vars)

`atomic_df`	Data frames with agent predictions.
`sotw`	Data frame containing the state of the world at each time point, which can include decision maker variables not in any of the atomic models. The first column of this data frame should be t (as in time).
`start_agg`	From which value of t to start aggregating, ie producing aggregate predictions.
`cw`	The caliper width.
`matching_vars`	Data frame with matching variables, ie pooling variables we want to fully match. First column should be t (and correspond in time to the other t columns).

A data table that consists of several stacked data tables (each indexed by the column t). Each subtable has a value t2 for each previous observation, and to each of those a corresponding similarity.

ooelrich/oscbvar documentation built on Sept. 8, 2021, 3:31 p.m.