pif.conditional.variance.linear: Conditional Variance for the Potential Impact Fraction

In pifpaf: Potential Impact Fraction and Population Attributable Fraction for Cross-Sectional Data

Description

Function that calculates the conditional variance of the potential impact fraction pif (linearization).

Usage

pif.conditional.variance.linear(X, thetahat, rr, cft = NA,
  weights = rep(1/nrow(as.matrix(X)), nrow(as.matrix(X))), check_cft = TRUE,
  is_paf = FALSE)

Arguments

X	Random sample (data.frame) which includes exposure and covariates.
thetahat	Estimator (vector) of theta for the Relative Risk function.
rr	function for Relative Risk which uses parameter theta. The order of the parameters shound be rr(X, theta). **Optional**
cft	Function cft(X) for counterfactual. Leave empty for the Population Attributable Fraction PAF where counterfactual is 0 exposure.
weights	Survey weights for the random sample X.
check_cft	Check if counterfactual function cft reduces exposure.
is_paf	Force evaluation of paf

Author(s)

Rodrigo Zepeda-Tello [e-mail]

Dalia Camacho-Garc<c3><ad>a-Forment<c3><ad> [e-mail]

See Also

pif.variance.linear for pif variance without conditioning on theta and pif.variance.loglinear for variance of log(pif)

Examples

#Example 1: Exponential Relative risk
#--------------------------------------------
set.seed(18427)
X        <- data.frame(rnorm(100,3,.5))
thetahat <- 0.12
rr       <- function(X, theta){exp(theta*X)}

#When no counterfactual is specified calculates PAF
pif.conditional.variance.linear(X, thetahat,  rr)

#Example with linear counterfactual
cft      <- function(X){0.3*X}
pif.conditional.variance.linear(X, thetahat,  rr = function(X, theta){exp(theta*X)}, cft)

#Example 2: Multivariate case
#--------------------------------------------
set.seed(18427)
X1 <- rnorm(100, 3,.5)
X2 <- runif(100, 1, 1.5)
X  <- data.frame(cbind(X1,X2))
thetahat  <- c(0.1, 0.03)
rr        <- function(X, theta){
           .X <- as.matrix(X, ncol = 2)
           exp(theta[1]*.X[,1] + theta[2]*.X[,2])
           }
cft <- function(X){0.5*X}
pif.conditional.variance.linear(X, thetahat, rr, cft) 

pifpaf documentation built on May 1, 2019, 9:11 p.m.