waifw-fitter: Fitting WAIFW Structures

In TeaKov/serostat: Modeling Infectious Disease Parameters Based on Serological and Social Contact Data R Package

Description

This function can be used to fit irregular WAIFW structures under constrained optimization. It requires the choice of WAIFW matrix as input (“waifw.choice”) and uses as starting values the parameters from waifw.6parms.

Usage

waifw.fitter(a, y, n, waifw.choice, muy, breaks, N, D, Lmax, A, startpar, plots)

Arguments

a	Numeric vector of ages.
y	The response variable (binary, 1=past infection, 0=otherwise).
n	Number of individuals.
waifw.choice	The choice of WAIFW structure.
muy	Age-specific mortality rates.
breaks	Breakpoints of the age-categories considered in the WAIFW approach.
N	Population size.
D	Mean duration of infectiousness.
Lmax	Maximum age.
A	Age of loss of maternal immunity (0<A<1).
startpar	Starting values for the “nlm” method used in this function.
plots	Generate plots during fitting process (TRUE or FALSE.

See Also

pwcrate, waifw.6parms

Examples

ND <- c(489,47,29,21,12,12,16,15,15,6,6,14,17,19,17,23,34,33,62,71,68,68,78,71,71,96,86,83,79,
  80,83,93,126,120,121,132,135,176,161,193,196,218,257,277,331,376,356,435,460,453,535,545,
  576,668,692,759,722,819,939,1015,1051,973,1113,996,940,1074,1252,1367,1468,1541,1661,1838,
  2012,2236,2517,2793,2938,2994,3311,3516,3727,3857,4088,4161,4261,4274,4061,2509,2049,2159,
  2205,2550,2330,1992,1569,1242,1000,726,533,996)
PS <- c(118366,117271,114562,113894,116275,118030,116761,117742,119583,119887,118963,119958,
  124637,129143,131030,129724,127187,126433,124377,124883,122201,124482,126459,130129,133897,
  135009,134516,133495,132705,132040,130602,135638,140537,146151,150467,152113,151656,151412,
  153371,158268,162456,167652,164871,161671,162060,159735,160672,157030,153820,151114,148978,
  145929,142374,141215,135525,135968,134692,135991,134291,134131,113024,112198,105880,92772,
  84462,93787,100820,101866,97208,94145,92451,93027,91640,93593,91933,89900,81718,77891,73104,
  70082,67057,62178,57642,51786,47466,42065,28004,17186,14492,13838,13957,13358,10442,8063,5604,
  4289,2843,2068,1368,2146)
AGE<-c(0:(length(ND)-1))

estimL <- estimateLifeExpectancy(ND, PS, AGE)

# Using Belgian B19 data
data("VZV_B19_BE_0103")
VZV_B19_BE_0103 <- VZV_B19_BE_0103[!is.na(VZV_B19_BE_0103$parvores)&
  !is.na(VZV_B19_BE_0103$age)&VZV_B19_BE_0103$age<70&VZV_B19_BE_0103$age>=1,]
VZV_B19_BE_0103 <- VZV_B19_BE_0103[order(VZV_B19_BE_0103$age),]

y <- VZV_B19_BE_0103$parvores
a <- VZV_B19_BE_0103$age

# Age category
breakpoints <- c(0.5,2,6,12,19,31,100)

# Mean duration of infectiousness
D <- 6/365

# Maximum life (if type mortality this is the life expectancy)
Lmax<-100

# Age of loss of maternal immunity (0<A<1)
A <- 0.5

# Mortality function
My <- estimL$My[1:Lmax]
muy <- estimL$muy[1:Lmax]

# Population size
N <- 10511382

# Fit WAIFW matrices W2 and W3
w2 <- waifw.fitter(a=a,y=y,waifw.choice=2,muy=muy,breaks=breakpoints,
  N=N,D=D,Lmax=Lmax,A=A,plots="TRUE")
w3 <- waifw.fitter(a=a,y=y,waifw.choice=3,muy=muy,breaks=breakpoints,
  N=N,D=D,Lmax=Lmax,A=A,plots="TRUE")

TeaKov/serostat documentation built on May 21, 2019, 1:21 p.m.