Skip to contents

Perform hierarchical moving-window statistical analysis for time series.

Source: R/TH_coupled.R
TH_coupled.Rd

Perform hierarchical moving-window statistical analysis for coupled time series. It returns analysis on moving correlations using Pearson's correlation coefficient and Mann-Kendall correlation coefficient. The results are in a data.frame to be passed to TH_plotc() function.

Usage

TH_coupled(
  series1,
  series2,
  m = NULL,
  s = NULL,
  alpha = 0.1,
  mode = "all",
  alternative = c("two.sided", "greater", "less")
)

Arguments

alternative

Optional, "two.sided" (Default), "greater", "less".

`series1`

a numerical vector containing NA for missing values.

`series2`

a numerical vector containing NA for missing values. It must be of the same length of series1.

`m`

a numeric positive integer, subsampling parameter. Optional (default: autocalculated).

`s`

a numeric positive integer, cutoff parameters for subseries legth. Optional (default: autocalculated).

`mode`

a character string specifying the computation mode. One of: "all", "pearson", "kendall", "both", "pearson_with_p", "kendall_with_p".

`alpha`

a numeric significance level for flagging (default: 0.1).

Value

A result data.frame to pass to TH_plotc() with attributes.

Author

Vladimiro Andrea Boselli, (2025) boselli.v@irea.cnr.it

Examples

 if (FALSE) {
output <- TH_coupled(
  series1 = rnorm(200),
  series2 = rnorm(200),
  m = 2,
  s = 6,
  alpha = 0.05,
  mode = "pearson_with_p",
  alternative = "less"
)
}