compute_offset_obs_vecs: Compute a data frame with offset (lag and/or lead)...

In reichlab/kcde: Kernel conditional density estimation with flexible kernel specifications

Description

Compute a data frame with offset (lag and/or lead) observation vectors

Usage

compute_offset_obs_vecs(data, filter_control, phi, vars_and_offsets, time_name,
  leading_rows_to_drop = 0, trailing_rows_to_drop = 0,
  additional_rows_to_drop = NULL, na.action = "na.omit")

Arguments

data	a data frame
vars_and_offsets	a named list: The component name matches the name of one of the variables in data, and the component value is an integer vector of lags to include for that variable
leading_rows_to_drop	an integer specifying the number of leading rows to drop. This will typically be the largest lag used, but could be larger for example if we are in the process of searching lags to determine which ones to include. Then for example if our search is for lags up to 52, we would want to set leading_rows_to_drop = 52.
trailing_rows_to_drop	an integer specifying the number of trailing rows to drop. This will typically be the largest prediction horizon used.
additional_rows_to_drop	an integer vector specifying indices of additional rows to drop. For example, if we are performing cross-validation, we might want to drop all rows within +/- 52 indices of the current prediction target.
rows_to_drop	an integer vector specifying rows to drop after computing lagged observation vectors.

reichlab/kcde documentation built on May 27, 2019, 4:53 a.m.