/wcm.gsa

WCM.gSa methodology for change point analysis in the presence of serial dependence

Primary LanguageR

wcm.gsa

Software accompanying

H. Cho and P. Fryzlewicz (2021) "Multiple change point detection under serial dependence: wild contrast maximisation and gappy Schwarz algorithm".

  • The main routine of WCM.gSa (wcm.gsa) is contained in main.R.
  • Codes used for analysing nitrogen oxides concentrations in London (contained in AirQualityData.csv) are provided in air_data.R.
  • Codes used for analysing Hadley Centre central England temperature data (contained in hadcet.csv) are provided in temp_data.R.

To use WCM.gSa, do the following:

  • Source main.R into R.
  • Read the descriptions for wcm.gsa within main.R.

For example,

source("main.R")
set.seed(111)
f <- rep(c(0, 5, 2, 8, 1, -2), c(100, 200, 200, 50, 200, 250))
x <- f + arima.sim(list(ar = c(.75, -.5), ma = c(.8, .7, .6, .5, .4, .3)), n = length(f), sd = 1)
wcm.gsa(x)

If you have any questions, contact haeran.cho@bristol.ac.uk