RRlog: Logistic randomized response regression

Description

A dichotomous variable, measured by a randomized response method, serves as dependent variable using one or more continuous and/or categorical predictors

Usage

RRlog(formula, data, model, p, group, LR.test = TRUE, fit.n = c(10, 100),
  fit.bound = 10, maxit = 1000, start = NULL, ...)

Arguments

formula

specifying the regression model, see formula

data

data.frame, in which variables can be found (optional)

model

Available RR models: "Warner","UQTknown","UQTunknown","Mangat","Kuk","FR","Crosswise","CDM","CDMsym","SLD". See vignette("RRreg") for details.

randomization probability/probabilities (depending on model)

group

vector specifying group membership. Can be omitted for single-group RR designs (e.g., Warner). For two-group RR designs (e.g., CDM or SLD), use 1 and 2 to indicate the group membership, matching the respective randomization proba

LR.test

test regression coefficients by a likelihood ratio test, i.e., fitting the model repeatedly while excluding one parameter at a time

fit.n

Minimum and maximum number of fitting replications using random starting values to avoid local maxima (only if start=NULL)

fit.bound

The model is fitted repeatedly either until the absolute parameter estimates are below fit.bound or the maximum number of fitting replication is reached. Thereby, stability of the estimates is increased. fit.bound should be incre

maxit

Maximum number of iterations within each run of optim

start

starting values for optimization. Might be useful if model does not converge with default starting values.

...

ignored

Value

Returns an object RRlog which can be analysed by the generic method summary

References

van den Hout, A., van der Heijden, P. G., & Gilchrist, R. (2007). The logistic regression model with response variables subject to randomized response. Computational Statistics & Data Analysis, 51, 6060-6069.

Examples

Run this code

# generate data set without biases
dat <- RRgen(1000,pi=.3,"Warner",p=.8)
dat$covariate <- rnorm(1000)
dat$covariate[dat$true==1] <- rnorm(sum(dat$true==1),.4,1)
# analyse
ana <- RRlog(response~covariate,dat,"Warner",.8, fit.n = c(1,5))
summary(ana)