get_st_indicators: Get spatio-temporal indicators from a data frame containing...

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

Description

Calculate spatio-temporal indicators of one or more variables within a spatio-temporal neighborhood of one or more maximum radius (in terms of spatio-temporal distance).

Usage

get_st_indicators(
  df,
  stations_sf,
  radiuses = c(0.1),
  stats = c("mean", "sd"),
  alpha = 0.5,
  neib_type = "cone",
  time_id = "time",
  site_id = "site_id",
  vars = c("value"),
  parallel = FALSE,
  nsplits = 4
)

Arguments

df	A data frame containing spatio-temporal information
stations_sf	An sf object containing geographical information on the location of df
radiuses	A vector of values defining the maximum spatio-temporal distance allowed for an observation to be considered within a spatio-temporal neighbourhood
stats	A vector containing the names of functions that are to be used to calculate summarizing statistics
alpha	a weighting factor for the spatio-temporal distance
neib_type	the type of neighborhood to consider. Can be cone (default) - a cone with the center of its base at the observation (spatial radius growing with time) reversed - a cone with its peak at the observation (spatial radius shrinking with time)
time_id	The name of the column containing time-stamps in df
site_id	The name of the column containing location IDs in df
vars	The name of the variables to summarize into indicators
parallel	Boolean indicating whether the code should run in parallel. Default is FALSE
nsplits	Number of subsets of rows to split the data frame into so they can be processed in parallel

Details

The spatio-temporal distance is defined as

D_{i,j} = d_{i,j} x α + t_{i,j} x (1-α)

where d_i,j is the spatial distance between locations, t_i,j is the temporal distance between time-stapms and α is a weighting factor.

Value

A data frame that contains summary statistics of the values of vars found within the spatio-temporal neighbourhoods of the one or more radiuses of each pair (location ID, time-stamp) in df

References

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.258.8742&rep=rep1&type=pdf

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