This package is an adaptation of the Matalb tsEVA toolbox developed by Lorenzo Mentaschi availaible here: https://github.com/menta78/tsEva
It contains an implementation of the Transformed-Stationary (TS) methodology for non-stationary EVA as described in Mentaschi et al. (2016). In synthesis this approach consists in (i) transforming a non-stationary time series into a stationary one to which the stationary EVA theory can be applied; and (ii) reverse-transforming the result into a non-stationary extreme value distribution.
You can install the released version of RtsEva from
CRAN with:
install.packages("RtsEva")Alternatively, you can install the development version of RtsEva from GitHub with:
# install.packages("devtools")
devtools::install_github("Alowis/RtsEva")This is a basic example which shows you how to solve a common problem:
library(RtsEva)
# Load a time series
timeAndSeries <- ArdecheStMartin
# go from six-hourly values to daily max
timeAndSeries <- max_daily_value(timeAndSeries)
# set a temporal window for the computation of running statistics
timeWindow <- 30*365 # 30 years
# Run the non-stationnary EVA
result <- TsEvaNs(timeAndSeries, timeWindow,
transfType = 'trendPeaks',tail = 'high')After fitting the non-stationnay EVA, the package offers functions to visualize the plots.
Other resources are available for users to get a grasp of RtsEva and what can be done with it:
A quarto document showing a step-by-step real world application of TSEVA for flood and drought analysis is available here: https://alowis.github.io/RTSEVA_guide/RtsEVA_demo.html
A shiny app allowing the user to select different input timeseries and parameters: https://alowis.shinyapps.io/RtsEva_demo/
https://alowis.quarto.pub/rtseva-package-demo/
For any questions or inquiries, please contact the package maintainer at [email protected]