CTRE (version 0.1.0)

CTRE-package: Continuous Time Random Exceedances

Description

Fits a Continuous Time Random Exceedance model to bursty time series. CTREs model the distribution of threshold crossing times of extreme events. Fits are made by interactive visual assessment of "stability plots" using a shiny app.

References

"Peaks Over Threshold for Bursty Time Series", Katharina Hees, Smarak Nayak, Peter Straka. https://arxiv.org/abs/1802.05218

See Also

Useful links:

• https://unsw-math.github.io/CTRE/

• Report bugs at https://github.com/UNSW-MATH/CTRE/issues

Examples

# NOT RUN {
library(CTRE)
# }
# NOT RUN {
runCTREshiny()
# }
# NOT RUN {
ctre_mod <- ctre(flares)
ctre_mod
length(ctre_mod)
ctre_mod <- thin(ctre_mod, k = 500)
ctre_mod
magnitudes(ctre_mod)
time(ctre_mod)
interarrival(ctre_mod)
MLestimates(ctre_mod, tail = 0.8, scale = 3E7)
library(magrittr)
ctre_mod %>% interarrival() %>% mlqqplot(tail = 0.8, log = 'xy')
acf(ctre_mod)
empcopula(ctre_mod)

# }