ThrSelection function

Threshold selection

This function selects a moving threshold for circular data using an automatic procedure for selecting the local number of exceedances

ThrSelection(
  Data,
  drc,
  h = 30,
  b = 0.35,
  thetaGrid,
  EVIestimator = "ML",
  useKernel = TRUE,
  concent = 10,
  bw = 30,
  numCores = 2
)

Arguments

• Data: Response variable
• drc: Directional covariate
• h: Bandwidth value
• b: Parameter used in the automatic procedure for selection of local number of exceedances
• thetaGrid: Grid values at which the estimation is performed
• EVIestimator: It can be either "ML" or "Mom"
• useKernel: Logical. If TRUE (default), use kernel to assign weights depending on the directional distance.
• concent: Concentration parameter value for von Mises kernel
• bw: Bandwidth parameter value for smoothing the sample path of the selected threshold
• numCores: Number of CPU cores to be used

Returns

List containing the selected threshold and selected number of local exceedances at each direction in the grid.

Details

See Konzen, E., Neves, C., and Jonathan, P. (2021). Modeling nonstationary extremes of storm severity: Comparing parametric and semiparametric inference. Environmetrics, 32(4), e2667.

Examples

data(HsSP)
data(drc)
timeRange <- 54.5

idx <- order(drc)
drc <- drc[idx]
Data <- HsSP[idx]
set.seed(1234)
Data <- Data + runif(length(Data), -1e-4, 1e-4)

thetaVec <- 1:360


thrResultMom <- ThrSelection(Data=Data, drc=drc, h=60, b=0.35, thetaGrid=thetaVec,
                             EVIestimator="Mom", useKernel=T, concent=10, bw=30,
                             numCores=2)$thr
thrResultML <- ThrSelection(Data=Data, drc=drc, h=60, b=0.35, thetaGrid=thetaVec,
                            EVIestimator="ML", useKernel=T, concent=10, bw=30,
                            numCores=2)$thr


## See also examples in vignettes:
# vignette("localMethods", package = "circularEV")
# vignette("splineML", package = "circularEV")

See Also

PlotData and PolarPlotData to see how the threshold can be visualised.