CCGamma: This function extends 'continuity_ratio' and adds the...

In ecor/RMRAINGEN: RMRAINGEN (R Multi-site RAINfall GENeretor): a package to generate daily time series of rainfall from monthly mean values

Description

This function extends continuity_ratio and adds the corresponding gaussian correlation matrix for no-precipitation occurence.

Usage

CCGamma(data, lag = 0, p0_v1 = NULL, p = NA, valmin = 0.5,
  nearPD = (lag >= 0), interval = c(-1, 1),
  tolerance = .Machine$double.eps, only.matrix = FALSE,
  return.value = NULL, null.gcorrelation = 1e-05, sample = NULL,
  origin = "1961-1-1", ...)

Arguments

data	data frame or 'zoo' R object containing daily precipitation time series for several gauges (one gauge time series per column). See continuity_ratio.
lag	numeric lag (expressed as number of days) used for computation for "cross" continuity ratio and joint probability of prercipitation (no)occurence. See continuity_ratio.
p	positive integer parameter. Default is NA, otherwise, lag is calculated as the vector 0:p.
valmin	threshold precipitation value [mm] for wet/dry day indicator. If precipitation is lower than valmin, day is considered dry. Default is 0.5 mm. See continuity_ratio.
p0_v1	vector for marginal probablities, see omega and omega_inv.
nearPD	see omega_inv. Default is (lag==0).
interval,tolerance	see omega_inv
only.matrix	logical value. If TRUE the function returns only the gaussian correlaton matrix. Deafaul is FALSE.
return.value	string. If it is not either NULL (Default) and NA, function returns only the argument indicated by this argument.
null.gcorrelation	numerical value nooccurence_gcorrelation under which is considered to be 0.
sample	character string indicated if function must be calculated differently for subset of the year, e.g. monthly. Admitted values are NULL (Default), "all" or "monthly".
origin	character string (yyyy-dd-mm) indicated the date of the first row of "data". It is used if data and sample are not NULL.
...	additional agruments of omega_inv or CCGamma

Value

An object which is a list containing the following fields:

continuity_ratio : lag-day lagged continuity ratio, as returned by continuity_ratio;

occurence : joint probability of lag-day lagged precipitation occurence, as returned by continuity_ratio;

nooccurence : joint probability of lag-day lagged no precipitation occurence, as returned by continuity_ratio;

lag : number of days lagged between the two compared events (see argument lag);

p0_v1 : vector of marginal probability of no precipitation occurence. If lag is 0, it corresponds to the diagonal of nooccurence matrix (see argument p0_v1);

nooccurence_gcorrelation corresponding gaussian correlation for no precipitation occurence obtained by applying omega_inv to nooccurence,

TransintionMatrixMCFirstOrder List of transition matrices (with condinitioned probability values) of the first-order Markov Chain. See TransitionMatrixMCFirstOrder.

If the argument only.matrix is TRUE, only nooccurence_gcorrelation is returned as a matrix. In case the argument lag is a vector wirh length more than one, the function returns a list of the above-cited return object for each value of the vector lag.

Note

This functon is useful to generate the serial cross-correlation matrices for no precipitation occurrence for Yule-Walker Equations. In case lag is a vactor, nearPD must be a vector of the same size, default is (lag==0).

See the R code for major details

Author(s)

Emanuele Cordano

See Also

continuity_ratio,omega_inv,omega,CCGammaToBlockmatrix

Examples

library(RMAWGEN)
data(trentino)

year_min <- 1961
year_max <- 1990
origin <- paste(year_min,1,1,sep="-")

period <- PRECIPITATION$year>=year_min & PRECIPITATION$year<=year_max
station <- names(PRECIPITATION)[!(names(PRECIPITATION) %in% c("day","month","year"))]
prec_mes <- PRECIPITATION[period,station]

## removing nonworking stations (e.g. time series with NA)
accepted <- array(TRUE,length(names(prec_mes)))
names(accepted) <- names(prec_mes)
for (it in names(prec_mes)) {
		 accepted[it]  <- (length(which(!is.na(prec_mes[,it])))==length(prec_mes[,it]))
}

prec_mes <- prec_mes[,accepted]
## the dateset is reduced!!!
prec_mes <- prec_mes[,1:2]

CCGamma <- CCGamma(data=prec_mes,lag=0,tolerance=0.001,only.matrix=FALSE)
## Not Run in the examples, uncomment to run the following line
## CCGamma <- CCGamma(data=prec_mes,lag=0:2,tolerance=0.001,only.matrix=FALSE)

## Not Run in the examples, uncomment to run the following line
## CCGamma_monthly <- CCGamma(data=prec_mes,lag=0,tolerance=0.001,only.matrix=FALSE,
#                     sample="monthly",origin=origin)

ecor/RMRAINGEN documentation built on May 15, 2019, 8:53 p.m.