Skip to contents

Run STL decomposition and trend tests on grouped time series data

Source: R/trend_analysis.R
run_grouped_stl_trend.Rd

Performs STL decomposition, Mann-Kendall trend testing, and Sen's slope estimation for multiple indicators within grouped time series data.

Usage

run_grouped_stl_trend(
  data,
  group_col,
  date_col,
  indicators,
  freq = 12,
  normalize_fun = normalize_zscore,
  stl_window = "periodic"
)

Arguments

data

A data.frame containing the time series data.

group_col

Character vector of grouping column names.

date_col

Name of the date column.

indicators

A list of indicator specifications. Each element must contain:

  • col: column name of the indicator

  • type: indicator type label

freq

Number of observations per year. Default is 12.

normalize_fun

Function used to normalize indicator values.

stl_window

STL seasonal window. Default is "periodic".

Value

A data.table containing STL components and trend statistics.

Examples

if (FALSE) { # \dontrun{
res <- run_grouped_stl_trend(
  data = monthly_adm2_incid,
  group_col = c("adm1", "adm2"),
  date_col = "date",
  indicators = indicators
)
} # }