double_cox_longevity2: double_cox_longevity2

In ilyasstatistics/mylongevity: Calculation of life expectancy (Calculation of life expectancy)

Description

This function loads a file as a data frame of clients and fits parametric survival model

Usage

double_cox_longevity2(
  data,
  dist,
  cluster,
  formula.shape,
  formula.scale,
  age_of_diagnosis,
  time_past_from_diagnosis,
  working_directory,
  name_for_age_factor = NULL
)

Arguments

data	given data frame of clients
dist	distribution either Weibull or Gompertz
cluster	The name of the covariate defining the random effect (is equal to NULL for the fixed-effect model);
formula.shape	The object defining the fields for covariates influencing the shape.
formula.scale	The formula object defining the fields for time-to-failure (or time-of-start and time-to-failure) and for covariates influencing the proportional hazard term;
age_of_diagnosis	age of diagnosis for a particular condition
time_past_from_diagnosis	time past diagnosis
working_directory	the working directory for any data that is produced from the functions to be saved in a folder
name_for_age_factor	the name of continuous age covariate

Details

This method fits the Double Cox regression to the provided data and using the hazard ratios from scale and shape parameters it calculates the life expectancies for all combination of given categorical variables

Value

data frame with life expectancies for given data frame of clients

Examples

#set.seed(1)
#n<-10000
#clinic<-as.factor(round(runif(n, min = 1, max = 1000)))###suppose we have 200 clinic
#gender <- c(rep('M',times=n/2),rep('F',times=n/2))
#townsend <- as.factor(round(runif(n, min = 1, max = 5)))
#smokerCategory <- as.factor(round(runif(n, min = 1, max = 3)))
#HTN_diag_treat  <- as.factor(round(runif(n, min = 1, max = 3)))
#diabetes  <- as.factor(round(runif(n, min = 0, max = 1)))
#hypercholesterolaemia <- as.factor(round(runif(n, min = 0, max = 1)))
#bmiCategory <- as.factor(round(runif(n, min = 1, max = 3)))
#age_cat<- as.factor(round(runif(n, min = 1, max = 3)))
#birth_cohort<- as.factor(round(runif(n, min = 0, max = 3)))
#cvd_risk <- as.factor(round(runif(n, min = 0, max = 2)))
#statins <- as.factor(round(runif(n, min = 0, max = 1)))
#aspirin<-as.factor(round(runif(n, min = 0, max = 1)))
#age_c<-rnorm(n,mean=70, sd=10)
#Tstart<-rep(60, times=n)
#Tstop<-abs(rnorm(n,mean=70,sd= 6))
#death<-sample(c(0,1), replace=TRUE, size=n)
#T_start_indicator <- "Tstart"
#T_stop_indicator <-  "Tstop"
#status_indicator <-  "death"
#working_directory<-"E:/Documentation for R package/"
#data<-data.frame(clinic,statins,age_c,birth_cohort,cvd_risk,diabetes,HTN_diag_treat,hypercholesterolaemia,bmiCategory,age_cat,smokerCategory,townsend,aspirin,Tstart,Tstop,death)
#data <- subset(data, Tstop>Tstart) #removing observations where Tstop happened before Tstart
#age_of_diagnosis<-60 ##age at diagnosis
#time_past_from_diagnosis<-5 ## time from diagnosis
#cluster="clinic"
#dist="Weibull"
#formula.scale=as.formula("survival::Surv(Tstart,Tstop,death)~birth_cohort+age_c+HTN_diag_treat+smokerCategory")
#formula.shape=as.formula("survival::Surv(Tstart,Tstop, death) ~birth_cohort")
#double_cox_longevity2(data, dist, cluster, formula.shape, formula.scale,age_of_diagnosis,time_past_from_diagnosis,working_directory,name_for_age_factor="age_c")

ilyasstatistics/mylongevity documentation built on Aug. 28, 2021, 4:51 p.m.