RandomThinning-package: Random thinning tests and auxiliary functions

In JeremySilver/RandomThinning: Random Thinning Tests and Auxiliary Functions

Description

Monte Carlo tests for weak signals in time-series

Details

This package provides the functions randomThinningTest (which performs the test in question), calculateTSstatistics (which calculates relevant properties of the time-series), calculatePower (which estimates the power of the test for a given data-set) and rangeTest (the statistic underlying these tests).

Author(s)

Jeremy Silver [e-mail]

References

These tests first appeared in Earl et al. 2015:

Earl, N., Simmonds, I. & Tapper, N. (2015) Weekly cycles of global fires - Associations with religion, wealth and culture, and insights into anthropogenic influences on global climate. Geophysical Research Letters. 42 (21): 9579–9589

Examples

## Generate a random time-series:
##  - length = 1000
##  - signal period = 8
##  - signal to noise ratio = 0.2
set.seed(42)

x <- rep(rnorm(8)*0.2,length.out = 1000) + rnorm(1000)
randomThinningTest(ts = x, p = 8, n = 1000,
                   fr = 0.025, returnExtraInfo = TRUE)

## compare with periods either too short or too long:
randomThinningTest(ts = x, p = 7, n = 1000, fr = 0.025)
randomThinningTest(ts = x, p = 9, n = 1000, fr = 0.025)

JeremySilver/RandomThinning documentation built on May 30, 2019, 9:40 p.m.