NHMSAR (version 1.0)

MeanDurUnder: Mean Duration of sojourn under a treshold

Description

Plot the mean duration of sojourn under thresholds for an observed time series and a simulated one with respect to teh empirical cumulative distribution function (cdf). Confidence intervals are plotted too.

Usage

MeanDurUnder(data, data.sim, u, alpha = 0.05)

Arguments

data

observed (or reference) time series, array of dimension T*N.samples*1

data.sim

simulated time series, array of dimension T*N.sim*1. N.sim have to be K*N.samples with K large enough (for instance, K=100)

u

vector of thresholds

alpha

1-confidence level for confidence intervals. Default = 0.05

Value

• Returns a plot and a list including ..$F : empirical cdf of data for levels u ..$mdu.data : mean duration under levels u for data ..$F.sim : empirical cdf of simulations for levels u ..$mdu.sim : mean duration under levels u for simulations ..$CI : confidence intervals of mean duration under levels u for simulations

See Also

valid_all, MeanDurOver

Examples

data(meteo.data)
data = array(meteo.data$temperature,c(31,41,1)) 
k = 40
T = dim(data)[1]
N.samples = dim(data)[2]
d = dim(data)[3]
M = 2
order = 2
theta.init = init.theta.MSAR(data,M=M,order=order,label="HH")
mod.hh= NULL
mod.hh$theta = theta.init
mod.hh$theta$A = matrix(c(0.40,0.88,-.09,-.13),2,2)
mod.hh$theta$A0 = matrix(c(6.75,1.08),2,1)
mod.hh$theta$sigma = matrix(c(1.76,3.40),2,1)
mod.hh$theta$prior = matrix(c(0.37,0.63),2,1)
mod.hh$theta$transmat = matrix(c(0.82,0.09,0.18,0.91),2,2)
B.sim = 50*N.samples
Y0 = array(data[1:2,sample(1:dim(data)[2],B.sim,replace=TRUE),],c(2,B.sim,1))
Y.sim = simule.nh.MSAR(mod.hh$theta,Y0=Y0,T,N.samples=B.sim)
u = seq(min(data),max(data),length.out=30)
MeanDurUnder(data,Y.sim$Y,u)