pif.variance.linear: Approximate Variance for the Potential Impact Fraction

In INSP-RH/pif: Potential Impact Fraction and Population Attributable Fraction for Cross-Sectional Data

Description

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

Usage

pif.variance.linear(X, thetahat, rr, thetavar, cft = NA,
  weights = rep(1/nrow(as.matrix(X)), nrow(as.matrix(X))),
  check_thetas = TRUE, check_cft = TRUE, check_exposure = TRUE,
  nsim = 1000, 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**
thetavar	Estimator of variance of thetahat
cft	Function cft(X) for counterfactual. Leave empty for the Population Attributable Fraction paf where counterfactual is 0 exposure.
weights	Normalized survey weights for the sample X.
check_thetas	Checks that theta parameters are correctly inputed
check_cft	Check if counterfactual function cft reduces exposure.
check_exposure	Check that exposure X is positive and numeric
nsim	Number of simulations for estimation of variance
is_paf	Boolean to force paf evaluation.

Author(s)

Rodrigo Zepeda-Tello [e-mail]

Dalia Camacho-García-Formentí [e-mail]

See Also

pif.variance.approximate.linear for pif variance and pif.confidence for confidence intervals of pif

Examples

#Example 1: Exponential Relative risk
#--------------------------------------------
set.seed(18427)
X <- data.frame(rnorm(100,3,.5))
thetahat <- 0.12
thetavar <- 0.1
pif.variance.linear(X, thetahat, function(X, theta){exp(theta*X)}, 
                    thetavar, nsim = 100) 

#Same example with linear counterfactual
cft      <- function(X){0.3*X}
pif.variance.linear(X, thetahat, function(X, theta){exp(theta*X)}, 
                    thetavar, cft, nsim = 100) 
                    
#Example 2: Multivariate case
#--------------------------------------------
## Not run: 
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)
thetavar <- matrix(c(0.1, 0, 0, 0.05), byrow = TRUE, nrow = 2)
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.variance.linear(X, thetahat, rr, thetavar, cft, nsim = 100) 

## End(Not run)

INSP-RH/pif documentation built on May 7, 2019, 6:01 a.m.