R/prior_dengue_fxns.R
In predsci/DICE: DICE: Dynamics of Interacting Community Epidemics
Defines functions
calc.epi.cor
calc.sqldata.cor
calc.sqldata.dist
get.sql.da
Documented in
calc.sqldata.cor
calc.sqldata.dist
get.sql.da
##
## Functions related to Data Augmentation for Dengue: calculate correlation with past season, distance matrix, historic NULL model
## and actual aumentation
##
calc.epi.cor <- function(mydata = NULL, all_years_epi = NULL) {
# This is the season
my.year = mydata$year
nregion = mydata$fit$nregion
mod_level = mydata$mod_level
fit_level = mydata$fit_level
model_raw = mydata$model$raw
fit_raw = mydata$fit$raw
nregion = mydata$fit$nregions
nregion1 = nregion + 1
}
calc.sqldata.cor <- function(mydata= NULL, all_years_epi = NULL, cases = NULL, nfit = NULL) {
#' Calculate Pearson Correlation Matrix Between Seasons
#'
#' \code{calc.sqldata.cor} Calculates an nyears by nyears correlation matrix for dengue incidence
#' @param mydata A dataframe with all the available data for this \pkg{DICE} run
#' @param all_years_epi the epi data for all years
#' @param epi the epi data for current season
#' @param nfit The number of data points used when calculating the correlation
#' @param cases AN array with the entire time series of incidence
#' @return corrMAT An nyear x nyears matrix with the Pearson correlation values
#' @examples
#' calc.cor(mydata= mydata, all_years_epi = al_years_epi, cases = all_years_epi$model$epi, nfit = nfit)
#' calc.cor(mydata= mydata, all_years_epi = al_years_epi, cases = all_years_epi$fit$epi[,iregion], nfit = nfit)
year.vec = all_years_epi$years
years = unique(year.vec)
nyears = length(years)
nperiods = mydata$nperiods
nperiodsData = mydata$nperiodsData
nperiodsFit = mydata$nperiodsFit
if (mydata$cadence == 'Monthly') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$months == mydata$months[1])
cadence.vec = all_years_epi$months
cadence.start = mydata$months[1]
cadence = 'month'
}
if (mydata$cadence == 'Weekly') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$weeks == mydata$weeks[1])
cadence.vec = all_years_epi$weeks
cadence.start = mydata$weeks[1]
cadence = 'week'
}
if (mydata$cadence == 'Daily') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$weeks == mydata$days[1])
cadence.vec = all_years_epi$days
cadence.start = mydata$days[1]
cadence = 'day'
}
year.start = mydata$years[1]
istart = which(all_years_epi$years == year.start & cadence.vec == cadence.start)
nlines = length(all_years_epi$years)
# This is the epi data we will calculate correlation with
# Now walk back in time
corrMAT = array(NA, c(nyears, nyears))
colnames(corrMAT) = rownames(corrMAT) = years
for (iyear in rev(1:(nyears))) {
istart = which(all_years_epi$years == (year.start-nyears+iyear) & cadence.vec == cadence.start)
if(length(istart) == 0) next
iend = istart + nperiods -1
if(iend > nlines) iend = nlines
x = cases[istart:iend]
if (all(is.na(x))) next
for (jyear in rev(1:nyears)) {
if (iyear == jyear) next
istart = which(all_years_epi$years == (year.start-nyears+jyear) & cadence.vec == cadence.start)
if(length(istart) == 0) next
iend = istart + nperiods -1
if(iend > nlines) iend = nlines
y = cases[istart:iend]
if(all(is.na(y))) next
corrMAT[iyear,jyear] = cor(x[1:nfit], y[1:nfit], method = "pearson", use = "pairwise.complete.obs")
}
}
return(corrMAT)
}
calc.sqldata.dist <- function(mydata= NULL, all_years_epi = NULL, cases = NULL, nfit = NULL) {
#' Calculate Euler Distance Matrix Between Seasons
#'
#' \code{calc.dist} Calculates an nyears X nyears Euler distance matrix for the incidence profiles
#' @param mydata A dataframe with all the available data for this \pkg{DICE} run
#' @param all_years_epi the epi data for all years
#' @param nfit The number of data points used when calculating the correlation
#' @param cases AN array with the entire time series of incidence
#' @return distMAT An nyear x nyears matrix with Euler distance values
#' @examples
#' calc.sqldata.dist(mydata= mydata, all_years_epi = al_years_epi, cases = all_years_epi$model$epi, nfit = nfit)
#' calc.sqldata.distr(mydata= mydata, all_years_epi = al_years_epi, cases = all_years_epi$fit$epi[,iregion], nfit = nfit)
year.vec = all_years_epi$years
years = unique(year.vec)
nyears = length(years)
nperiods = mydata$nperiods
nperiodsData = mydata$nperiodsData
nperiodsFit = mydata$nperiodsFit
if (mydata$cadence == 'Monthly') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$months == mydata$months[1])
cadence.vec = all_years_epi$months
cadence.start = mydata$months[1]
cadence = 'month'
}
if (mydata$cadence == 'Weekly') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$weeks == mydata$weeks[1])
cadence.vec = all_years_epi$weeks
cadence.start = mydata$weeks[1]
cadence = 'week'
}
if (mydata$cadence == 'Daily') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$weeks == mydata$days[1])
cadence.vec = all_years_epi$days
cadence.start = mydata$days[1]
cadence = 'day'
}
year.start = mydata$years[1]
istart = which(all_years_epi$years == year.start & cadence.vec == cadence.start)
nlines = length(all_years_epi$years)
# This is the epi data we will calculate correlation with
# Now walk back in time
distMAT = array(NA, c(nyears, nyears))
colnames(distMAT) = rownames(distMAT) = years
for (iyear in rev(1:(nyears))) {
istart = which(all_years_epi$years == (year.start-nyears+iyear) & cadence.vec == cadence.start)
if(length(istart) == 0) next
iend = istart + nperiods -1
if(iend > nlines) iend = nlines
x = cases[istart:iend]
if (all(is.na(x))) next
for (jyear in rev(1:nyears)) {
istart = which(all_years_epi$years == (year.start-nyears+jyear) & cadence.vec == cadence.start)
if(length(istart) == 0) next
iend = istart + nperiods -1
if(iend > nlines) iend = nlines
y = cases[istart:iend]
if(all(is.na(y))) next
distMAT[iyear,jyear] = sqrt(sum((as.numeric(x[1:nfit] - y[1:nfit])) * (as.numeric(x[1:nfit] - y[1:nfit]))))/nfit
}
}
return(distMAT)
}
get.sql.da <- function(nfit = NULL, mydata = NULL, all_years_epi = NULL, cases = NULL, epi = NULL,
corrMAT = NULL, deng.null = NULL, my_id = NULL) {
#'
#' Data Augmentation for Mechanistic Model
#'
#' \code{get.sql.da} Allows the User to augment the incidence time series beyond the \code{nfit} data points
#' using either the historic NULL model \code{da = 1}, or the most similar past season \code{da=2}.
#' If \code{da=0}, the data is not augmented.
#' @param mydata A dataframe with all the available data for this \pkg{DICE} run
#' @param all_years_epi the epi data for all years
#' @param cases An array with the entire history of incidence information for one region/patch
#' @param epi An array with the incidence information for one season and one region/patch
#' @param nfit An integer - the number of data points use in the run
#' @param my_id The abbreviation of the state/region
#' @param corrMAT A 2D array with the Pearson corrletion values for all available disease seasons
#' @param deng.null the Historic NULL model - average monthly/weekly number of cases
#' @return epi A list with the original and augmented mydata (if da=1,2) and the weight for the augmented
#' data points
#' @examples
#' get.sql.da(nfit = nfit, mydata = mydata, years = all_years_epi$years, cases = all_years_epi$model$epi,
#' epi = mydata$model$epi,corrMAT = corrMAT, distMAT = distMAT, deng.null=deng.null)
#'
#' get.sql.da(nfit = nfit, mydata = mydata, years = all_years_epi$years, cases = all_years_epi$fit$epi[,iregion],
#' epi = mydata$fit$epi[,iregion],corrMAT = corrMAT, distMAT = distMAT, deng.null=deng.null)
if (is.null(nfit) | is.null(cases) | is.null(epi))
return
da = mydata$da
if (is.null(da))
da = 0
cadence = mydata$cadence
nperiods = mydata$nperiods
nperiodsFit = mydata$nperiodsFit
if (mydata$cadence == 'Monthly') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$months == mydata$months[1])
cadence.vec = all_years_epi$months
cadence.start = mydata$months[1]
cadence = 'month'
}
if (mydata$cadence == 'Weekly') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$weeks == mydata$weeks[1])
cadence.vec = all_years_epi$weeks
cadence.start = mydata$weeks[1]
cadence = 'week'
}
if (mydata$cadence == 'Daily') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$weeks == mydata$days[1])
cadence.vec = all_years_epi$days
cadence.start = mydata$days[1]
cadence = 'day'
}
augment = FALSE
if (da > 0)
augment = TRUE
epi.state = epi
nperiods = mydata$nperiods
nda = nperiods - nfit
if (nda == 0)
augment = FALSE
if (augment) {
cat("\n Entered da == TRUE Option \n\n")
epi.save = epi
epi.state.save = epi.state
my.year = mydata$season
print(my.year)
iyear = which(rownames(corrMAT) == my.year)
iyear = as.numeric(iyear)
if (all(is.na(corrMAT[iyear, ])) | da == 1) {
corr.max = -1
} else {
corr.max = max(corrMAT[iyear, ], na.rm = TRUE)
}
if (corr.max > 0) { ## Looking for the Most similar year
jyear = which.max(corrMAT[iyear, ])
year.start = mydata$years[1]
nyears = length(unique(all_years_epi$years))
istart = which(all_years_epi$years == (year.start-nyears+jyear) & cadence.vec == cadence.start)
#if(length(istart) == 0)
iend = istart + mydata$nperiods -1
nlines = length(cases)
if(iend > nlines) iend = nlines
y = cases[istart:iend]
futur = cases[istart:iend]
shft = epi.state[nfit] - futur[nfit]
x = futur[1:nfit]
y = epi.state[1:nfit]
shft = y[nfit] - x[nfit]
prob = cor(x, y, method = "pearson", use = "pairwise.complete.obs")
#b1 = sum((x - mean(x)) * (y - mean(y)))/sum((x - mean(x)) * (x - mean(x)))
#b0 = mean(y) - b1 * mean(x)
#z = b0 + b1 * futur
cat("\n Using Pearson Correlation to Choose Season for Data Augmentation \n")
cat("\n Season Selected is: ", colnames(corrMAT)[jyear], "\n")
} else {
cat(" Using Historic NULL model to Augment Data \n")
futur = deng.null[,my_id]
x = futur[1:nfit]
y = epi.state[1:nfit]
shft = epi.state[nfit] - futur[nfit]
prob = cor(x, y, method = "pearson", use = "pairwise.complete.obs")
#b1 = sum((x - mean(x)) * (y - mean(y)))/sum((x - mean(x)) * (x - mean(x)))
#b0 = mean(y) - b1 * mean(x)
#z = b0 + b1 * futur
}
epi.new = epi.state
wght = rep(1, nperiods)
if (nfit < nperiods) epi.new[(nfit + 1):nperiods] = futur[(nfit + 1):nperiods] + shft
if (prob <= 0 | length(prob) == 0) prob = (nperiods-nfit)/nperiods
if (nfit < nperiods) {
wght[(nfit + 1):nperiods] = prob ##cor(x[1:nfit], y[1:nfit])
epi.new[(nfit + 1):nperiods] = round(epi.new[(nfit + 1):nperiods])
}
for (i in (nfit + 1):nperiods) {
epi.new[i] = max(1, epi.new[i])
}
} else {
cat("\n\n NO Data Augmentation \n\n")
epi.save = epi
epi.state.save = epi.state
epi.new = epi.state
wght = rep(0, nperiods)
wght[1:nperiods] = 1
}
return(list(nfit = nfit, epi.new = epi.new, wght = wght))
}
predsci/DICE documentation built on Aug. 9, 2019, 9:41 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions calc.epi.cor calc.sqldata.cor calc.sqldata.dist get.sql.da
Documented in calc.sqldata.cor calc.sqldata.dist get.sql.da
##
## Functions related to Data Augmentation for Dengue: calculate correlation with past season, distance matrix, historic NULL model
## and actual aumentation
##
calc.epi.cor <- function(mydata = NULL, all_years_epi = NULL) {
# This is the season
my.year = mydata$year
nregion = mydata$fit$nregion
mod_level = mydata$mod_level
fit_level = mydata$fit_level
model_raw = mydata$model$raw
fit_raw = mydata$fit$raw
nregion = mydata$fit$nregions
nregion1 = nregion + 1
}
calc.sqldata.cor <- function(mydata= NULL, all_years_epi = NULL, cases = NULL, nfit = NULL) {
#' Calculate Pearson Correlation Matrix Between Seasons
#'
#' \code{calc.sqldata.cor} Calculates an nyears by nyears correlation matrix for dengue incidence
#' @param mydata A dataframe with all the available data for this \pkg{DICE} run
#' @param all_years_epi the epi data for all years
#' @param epi the epi data for current season
#' @param nfit The number of data points used when calculating the correlation
#' @param cases AN array with the entire time series of incidence
#' @return corrMAT An nyear x nyears matrix with the Pearson correlation values
#' @examples
#' calc.cor(mydata= mydata, all_years_epi = al_years_epi, cases = all_years_epi$model$epi, nfit = nfit)
#' calc.cor(mydata= mydata, all_years_epi = al_years_epi, cases = all_years_epi$fit$epi[,iregion], nfit = nfit)
year.vec = all_years_epi$years
years = unique(year.vec)
nyears = length(years)
nperiods = mydata$nperiods
nperiodsData = mydata$nperiodsData
nperiodsFit = mydata$nperiodsFit
if (mydata$cadence == 'Monthly') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$months == mydata$months[1])
cadence.vec = all_years_epi$months
cadence.start = mydata$months[1]
cadence = 'month'
}
if (mydata$cadence == 'Weekly') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$weeks == mydata$weeks[1])
cadence.vec = all_years_epi$weeks
cadence.start = mydata$weeks[1]
cadence = 'week'
}
if (mydata$cadence == 'Daily') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$weeks == mydata$days[1])
cadence.vec = all_years_epi$days
cadence.start = mydata$days[1]
cadence = 'day'
}
year.start = mydata$years[1]
istart = which(all_years_epi$years == year.start & cadence.vec == cadence.start)
nlines = length(all_years_epi$years)
# This is the epi data we will calculate correlation with
# Now walk back in time
corrMAT = array(NA, c(nyears, nyears))
colnames(corrMAT) = rownames(corrMAT) = years
for (iyear in rev(1:(nyears))) {
istart = which(all_years_epi$years == (year.start-nyears+iyear) & cadence.vec == cadence.start)
if(length(istart) == 0) next
iend = istart + nperiods -1
if(iend > nlines) iend = nlines
x = cases[istart:iend]
if (all(is.na(x))) next
for (jyear in rev(1:nyears)) {
if (iyear == jyear) next
istart = which(all_years_epi$years == (year.start-nyears+jyear) & cadence.vec == cadence.start)
if(length(istart) == 0) next
iend = istart + nperiods -1
if(iend > nlines) iend = nlines
y = cases[istart:iend]
if(all(is.na(y))) next
corrMAT[iyear,jyear] = cor(x[1:nfit], y[1:nfit], method = "pearson", use = "pairwise.complete.obs")
}
}
return(corrMAT)
}
calc.sqldata.dist <- function(mydata= NULL, all_years_epi = NULL, cases = NULL, nfit = NULL) {
#' Calculate Euler Distance Matrix Between Seasons
#'
#' \code{calc.dist} Calculates an nyears X nyears Euler distance matrix for the incidence profiles
#' @param mydata A dataframe with all the available data for this \pkg{DICE} run
#' @param all_years_epi the epi data for all years
#' @param nfit The number of data points used when calculating the correlation
#' @param cases AN array with the entire time series of incidence
#' @return distMAT An nyear x nyears matrix with Euler distance values
#' @examples
#' calc.sqldata.dist(mydata= mydata, all_years_epi = al_years_epi, cases = all_years_epi$model$epi, nfit = nfit)
#' calc.sqldata.distr(mydata= mydata, all_years_epi = al_years_epi, cases = all_years_epi$fit$epi[,iregion], nfit = nfit)
year.vec = all_years_epi$years
years = unique(year.vec)
nyears = length(years)
nperiods = mydata$nperiods
nperiodsData = mydata$nperiodsData
nperiodsFit = mydata$nperiodsFit
if (mydata$cadence == 'Monthly') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$months == mydata$months[1])
cadence.vec = all_years_epi$months
cadence.start = mydata$months[1]
cadence = 'month'
}
if (mydata$cadence == 'Weekly') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$weeks == mydata$weeks[1])
cadence.vec = all_years_epi$weeks
cadence.start = mydata$weeks[1]
cadence = 'week'
}
if (mydata$cadence == 'Daily') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$weeks == mydata$days[1])
cadence.vec = all_years_epi$days
cadence.start = mydata$days[1]
cadence = 'day'
}
year.start = mydata$years[1]
istart = which(all_years_epi$years == year.start & cadence.vec == cadence.start)
nlines = length(all_years_epi$years)
# This is the epi data we will calculate correlation with
# Now walk back in time
distMAT = array(NA, c(nyears, nyears))
colnames(distMAT) = rownames(distMAT) = years
for (iyear in rev(1:(nyears))) {
istart = which(all_years_epi$years == (year.start-nyears+iyear) & cadence.vec == cadence.start)
if(length(istart) == 0) next
iend = istart + nperiods -1
if(iend > nlines) iend = nlines
x = cases[istart:iend]
if (all(is.na(x))) next
for (jyear in rev(1:nyears)) {
istart = which(all_years_epi$years == (year.start-nyears+jyear) & cadence.vec == cadence.start)
if(length(istart) == 0) next
iend = istart + nperiods -1
if(iend > nlines) iend = nlines
y = cases[istart:iend]
if(all(is.na(y))) next
distMAT[iyear,jyear] = sqrt(sum((as.numeric(x[1:nfit] - y[1:nfit])) * (as.numeric(x[1:nfit] - y[1:nfit]))))/nfit
}
}
return(distMAT)
}
get.sql.da <- function(nfit = NULL, mydata = NULL, all_years_epi = NULL, cases = NULL, epi = NULL,
corrMAT = NULL, deng.null = NULL, my_id = NULL) {
#'
#' Data Augmentation for Mechanistic Model
#'
#' \code{get.sql.da} Allows the User to augment the incidence time series beyond the \code{nfit} data points
#' using either the historic NULL model \code{da = 1}, or the most similar past season \code{da=2}.
#' If \code{da=0}, the data is not augmented.
#' @param mydata A dataframe with all the available data for this \pkg{DICE} run
#' @param all_years_epi the epi data for all years
#' @param cases An array with the entire history of incidence information for one region/patch
#' @param epi An array with the incidence information for one season and one region/patch
#' @param nfit An integer - the number of data points use in the run
#' @param my_id The abbreviation of the state/region
#' @param corrMAT A 2D array with the Pearson corrletion values for all available disease seasons
#' @param deng.null the Historic NULL model - average monthly/weekly number of cases
#' @return epi A list with the original and augmented mydata (if da=1,2) and the weight for the augmented
#' data points
#' @examples
#' get.sql.da(nfit = nfit, mydata = mydata, years = all_years_epi$years, cases = all_years_epi$model$epi,
#' epi = mydata$model$epi,corrMAT = corrMAT, distMAT = distMAT, deng.null=deng.null)
#'
#' get.sql.da(nfit = nfit, mydata = mydata, years = all_years_epi$years, cases = all_years_epi$fit$epi[,iregion],
#' epi = mydata$fit$epi[,iregion],corrMAT = corrMAT, distMAT = distMAT, deng.null=deng.null)
if (is.null(nfit) | is.null(cases) | is.null(epi))
return
da = mydata$da
if (is.null(da))
da = 0
cadence = mydata$cadence
nperiods = mydata$nperiods
nperiodsFit = mydata$nperiodsFit
if (mydata$cadence == 'Monthly') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$months == mydata$months[1])
cadence.vec = all_years_epi$months
cadence.start = mydata$months[1]
cadence = 'month'
}
if (mydata$cadence == 'Weekly') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$weeks == mydata$weeks[1])
cadence.vec = all_years_epi$weeks
cadence.start = mydata$weeks[1]
cadence = 'week'
}
if (mydata$cadence == 'Daily') {
my_row = which(all_years_epi$years == mydata$years[1] & all_years_epi$weeks == mydata$days[1])
cadence.vec = all_years_epi$days
cadence.start = mydata$days[1]
cadence = 'day'
}
augment = FALSE
if (da > 0)
augment = TRUE
epi.state = epi
nperiods = mydata$nperiods
nda = nperiods - nfit
if (nda == 0)
augment = FALSE
if (augment) {
cat("\n Entered da == TRUE Option \n\n")
epi.save = epi
epi.state.save = epi.state
my.year = mydata$season
print(my.year)
iyear = which(rownames(corrMAT) == my.year)
iyear = as.numeric(iyear)
if (all(is.na(corrMAT[iyear, ])) | da == 1) {
corr.max = -1
} else {
corr.max = max(corrMAT[iyear, ], na.rm = TRUE)
}
if (corr.max > 0) { ## Looking for the Most similar year
jyear = which.max(corrMAT[iyear, ])
year.start = mydata$years[1]
nyears = length(unique(all_years_epi$years))
istart = which(all_years_epi$years == (year.start-nyears+jyear) & cadence.vec == cadence.start)
#if(length(istart) == 0)
iend = istart + mydata$nperiods -1
nlines = length(cases)
if(iend > nlines) iend = nlines
y = cases[istart:iend]
futur = cases[istart:iend]
shft = epi.state[nfit] - futur[nfit]
x = futur[1:nfit]
y = epi.state[1:nfit]
shft = y[nfit] - x[nfit]
prob = cor(x, y, method = "pearson", use = "pairwise.complete.obs")
#b1 = sum((x - mean(x)) * (y - mean(y)))/sum((x - mean(x)) * (x - mean(x)))
#b0 = mean(y) - b1 * mean(x)
#z = b0 + b1 * futur
cat("\n Using Pearson Correlation to Choose Season for Data Augmentation \n")
cat("\n Season Selected is: ", colnames(corrMAT)[jyear], "\n")
} else {
cat(" Using Historic NULL model to Augment Data \n")
futur = deng.null[,my_id]
x = futur[1:nfit]
y = epi.state[1:nfit]
shft = epi.state[nfit] - futur[nfit]
prob = cor(x, y, method = "pearson", use = "pairwise.complete.obs")
#b1 = sum((x - mean(x)) * (y - mean(y)))/sum((x - mean(x)) * (x - mean(x)))
#b0 = mean(y) - b1 * mean(x)
#z = b0 + b1 * futur
}
epi.new = epi.state
wght = rep(1, nperiods)
if (nfit < nperiods) epi.new[(nfit + 1):nperiods] = futur[(nfit + 1):nperiods] + shft
if (prob <= 0 | length(prob) == 0) prob = (nperiods-nfit)/nperiods
if (nfit < nperiods) {
wght[(nfit + 1):nperiods] = prob ##cor(x[1:nfit], y[1:nfit])
epi.new[(nfit + 1):nperiods] = round(epi.new[(nfit + 1):nperiods])
}
for (i in (nfit + 1):nperiods) {
epi.new[i] = max(1, epi.new[i])
}
} else {
cat("\n\n NO Data Augmentation \n\n")
epi.save = epi
epi.state.save = epi.state
epi.new = epi.state
wght = rep(0, nperiods)
wght[1:nperiods] = 1
}
return(list(nfit = nfit, epi.new = epi.new, wght = wght))
}
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.