merge_ppgrid: Merge a 'ppgrid' object to new data.

In DexGroves/ppR: Make prior period summary variables from longitudinal data.

Description

Find the just-in-time summarised response and weight for new data as informed by a ppgrid object. As train_ppgrid, merge_ppgrid expects ordered vectors.

Usage

merge_ppgrid(id_var, time_var, ppgrid, na_fill = FALSE)

Arguments

id_var	Vector identifying the groups.
time_var	Numeric vector of time, there is no support for character time vectors or other datetime datatypes yet.
ppgrid	ppgrid object produced by train_ppgrid.
na_fill	Boolean: whether to merge the furthest in the future values from train_ppgrid in the event that the new data's times are more advanced than the trained range.

Value

list containing the merged mean response and total weight for the input after merging with ppgrid. Ordered as input vectors.

Examples

set.seed(1234)
ppdf <- make_longitudinal_data(1000)
ppdf_train <- ppdf[ppdf$date < 1150, ]
ppdf_test <- ppdf[ppdf$date >= 1150, ]
ppgrid <- train_ppgrid(ppdf_train$id,
                       ppdf_train$date,
                       ppdf_train$resp,
                       lag = 25,
                       window_size = 25,
                       granularity = 25)

merge_ppgrid(ppdf_test$id,
             ppdf_test$date,
             ppgrid)

DexGroves/ppR documentation built on May 6, 2019, 2:13 p.m.