Disagg_help: Disaggregation of a single value to a timeseries sequence...

In itsoukal/anySim: Stochastic Simulation of Processes with any Marginal Distribution and Correlation Structure

Description

Disaggregation of a single value to a timeseries sequence exhibiting the target marginal distribution and correlation structure (stationary).

Usage

Disagg_help(HLValue, Zprevious, ARTApar, max.iter, steps, Adjust = T)

Arguments

HLValue	A scalar specifying the (single) value to disaggreate into a time series sequence.
ARTApar	A list containing the parameters of the model. The list is constructed by the function "EstARTAp".
max.iter	A scalar specifying the maximum number of allowed repetitions (parameter of the disaggregation algorithm - typically set between 300-500.).
steps	A scalar specifying the number of timesteps of the sequence to generate.
Adjust	A logical operator (TRUE or FALSE) specifying whether (TRUE) or not (FALSE) to perfom the proportianal adjusting operation (parameter of the disaggregation algorithm - typically set to TRUE).

Value

A list of the 3 generated time series (in vector format): X: The final time series at the actual domain with the target marginal distribution and correlation structure; Z: The auxiliary Gaussian time series at the Gaussian domain and; Diff: The difference between the (single) value to disaggreate with the sum of the generated sequence (i.e., aggregated to the same temporal level with the target value). Dicrit: The sum of the generated sequence (i.e., aggregated to the same temporal level with the target value).

Examples

## Disaggregation of a single 24-hour Rainfall ammount to 1-minute sequence.
## The lower level process (i.e., that of 1-minute) is assumed to be a zero-inflated one
## (i.e., using a mixed marginal distribution), with p0=0.95, and a Gamma distribution
## for the continuous part with shape=1 and scale=8.
## In this case, the target autocorrelation strucure is from
## the CAS ACS with b=0 and k=0.7.
## Not run: 
FX='qmixed'
PFX=list(p0=0.95, Distr=qgamma, shape=1, scale=.8)
maxlag=60
ACFT=acsCAS(param = c(0, 0.2), lag=maxlag)
param=EstARTAp(ACF = ACFT, dist = FX, params = PFX, NatafIntMethod = 'GH')

Sim=SimARTAp(ARTApar = param, burn = 1000, steps = 24*60*100) #24h*60min*100Days

SUMX=apply(X = matrix(data = Sim$X, ncol=24*60, byrow = 0), MARGIN = 1, FUN = sum)
HLValue=as.vector(quantile(SUMX, probs = 0.5, type = 0));HLValue
disag=Disagg_SV_S(HLValue = HLValue, ARTApar = param,
                          max.iter = 300, steps = 24*60, Adjust = 1)$X

## End(Not run)

itsoukal/anySim documentation built on May 7, 2020, 11:57 p.m.