Clean speed outliers using data-driven threshold

Usage

grz_clean_speed_stat(
  data,
  method = c("mad", "quantile"),
  k = 4,
  prob = 0.995,
  min_threshold_mps = 4,
  groups = NULL,
  keep_speed_cols = FALSE,
  verbose = TRUE,
  snapshot = FALSE,
  return_class = c("data.frame", "data.table")
)

Arguments

data: Data frame of GPS rows.
method: Threshold method: "mad" or "quantile".
k: MAD multiplier (used when method = "mad").
prob: Quantile probability (used when method = "quantile").
min_threshold_mps: Lower bound for threshold.
groups: Grouping columns for step/speed calculation.
keep_speed_cols: Keep step_dt_s, step_m, and speed_mps columns.
verbose: Logical; print drop counts.
snapshot: Logical; print quick snapshot after step.
return_class: Output class: "data.frame" (default) or "data.table".

Value

Cleaned data.