# POTevents.fun: Calculate extreme events using a POT approach In NHPoisson: Modelling and Validation of Non Homogeneous Poisson Processes

## Description

This function calculates the characteristics of the extreme events of a series (x_i) defined using a peak over threshold (POT) method with an extreme threshold. The initial and the maximum intensity positions, the mean excess, the maximum excess and the length of each event are calculated.

## Usage

 1 POTevents.fun(T, thres, date = NULL) 

## Arguments

 T Numeric vector, the series (x_i) to calculate the extreme events. thres Threshold value used to define the extreme events. date Optional. A vector or matrix indicating the date of each observation.

## Details

One of the elements of the output from this function is a vector (inddat) which marks the observations that should be used in the estimation of a point process, resulting from a POT approach. The observations to be considered in the estimation are marked with 1 and correspond to the non occurrence observations and to a single occurrence point per event. The occurence point is defined as the point where maximum intensity of the event occurs.The observations in an extreme event which are not the occurrence point are marked with 0 and treated as non observed.

## Value

A list with components

 Pi  Vector of the initial points of the extreme events. datePi  Date of the initial points Pi. Px  Vector of the points of maximum excess of the extreme events. datePx  Vector of the date of the maximum excess points Px. Im  Vector of the mean excesses (over the threshold) of the extreme events. Ix  Vector of the maximum excesses (over the threshold) of the extreme events. L  Vector of the lengths of the extreme events. inddat  Index equal to 1 in the observations used in the estimation process and to 0 in the others.

## See Also

fitPP.fun

## Examples

 1 2 3 data(BarTxTn) dateB<-cbind(BarTxTn$ano,BarTxTn$mes,BarTxTn$diames) BarEv<-POTevents.fun(T=BarTxTn$Tx,thres=318, date=dateB) 

### Example output

Number of events:  137
Number of excesses over threshold 318 : 253


NHPoisson documentation built on Feb. 19, 2020, 5:07 p.m.