sample_wts: Get spatio-temporal re-sampling weights

In mrfoliveira/STResampling-JDSA2020: Biased Resampling Strategies for Imbalanced Spatio-Temporal Forecasting

Description

A function that calculates different weights for re-sampling that is temporally and/or spatially biased.

Usage

sample_wts(
  form,
  df,
  phi.control,
  site_id,
  time,
  rel.thr,
  ptype,
  alpha,
  sites_sf = NULL,
  lon = NULL,
  lat = NULL,
  crs = NULL
)

Arguments

form	a formula describing the learning task
df	a data frame
phi.control	list of parameters as returned by function UBL::phi.control
site_id	the name of the column containing location IDs
time	the column name of the time-stamp
rel.thr	a relevance threshold above which an observation is considered relevant
ptype	type of spatio-temporal bias to calculate
alpha	weighting parameter for temporal and spatial re-sampling probabilities. Default 0.5
sites_sf	An sf obejct containing station and IDs and geometry points of the locations. As an alternative, provide lon, lat, and crs
lon	the name of the column containing the location's longitude
lat	the name of the column containing the location's latitude
crs	the code for the Coordinate Reference System

Details

phi gives the target variable's relevance (higher relevance: phi -> 1; lower relevance: phi -> 0); time_wts gives the observation's temporally biased re-sampling weight (most recent observations: w -> 1; oldest: w -> 0.); space_wts gives the observation's spatially biased re-sampling weight (farthest away from other relevant cases at time slice: d -> 1.). High time_wts or space_wts means the observation is more likely to be kept.

Value

a data.frame with relevance phi, temporally biased weights time_wts, and spatially biased weights space_wts for each row in df.

See Also

get_phi, get_time_wts, get_phi_space_wts.

mrfoliveira/STResampling-JDSA2020 documentation built on June 28, 2021, 7:01 p.m.