predict.rrreg.predictor
is used to generate predicted probabilities
from a multivariate regression object of survey data using the randomized
response item as a predictor for an additional outcome.
predict.rrreg.predictor(object, fix.z = NULL, alpha = .05, n.sims = 1000, avg = FALSE, newdata = NULL, quasi.bayes = FALSE, keep.draws = FALSE, ...)
object  An object of class "rrreg.predictor" generated by the


fix.z  An optional value or vector of values between 0 and 1 that the
user inputs as the proportion of respondents with the sensitive trait or
probability that each respondent has the sensitive trait, respectively. If
the user inputs a vector of values, the vector must be the length of the
data from the "rrreg.predictor" object. Default is 
alpha  Confidence level for the hypothesis test to generate upper and
lower confidence intervals. Default is 
n.sims  Number of sampled draws for quasibayesian predicted
probability estimation. Default is 
avg  Whether to output the mean of the predicted probabilities and
uncertainty estimates. Default is 
newdata  Optional new data frame of covariates provided by the user. Otherwise, the original data frame from the "rreg" object is used. 
quasi.bayes  Option to use Monte Carlo simulations to generate
uncertainty estimates for predicted probabilities. Default is 
keep.draws  Option to return the Monte Carlos draws of the quantity of interest, for use in calculating differences for example. 
...  Further arguments to be passed to

predict.rrreg.predictor
returns predicted probabilities
either for each observation in the data frame or the average over all
observations. The output is a list that contains the following components:
Predicted probabilities of the additional outcome variable given
the randomized response item as a predictor generated either using fitted
values or quasiBayesian simulations. If avg
is set to TRUE
,
the output will only include the mean estimate.
Standard errors
for the predicted probabilities of the additional outcome variable given the
randomized response item as a predictor generated using Monte Carlo
simulations. If quasi.bayes
is set to FALSE
, no standard
errors will be outputted.
Estimates for the lower
confidence interval. If quasi.bayes
is set to FALSE
, no
confidence interval estimate will be outputted.
Estimates
for the upper confidence interval. If quasi.bayes
is set to
FALSE
, no confidence interval estimate will be outputted.
Monte Carlos draws of the quantity of interest, returned
only if keep.draws
is set to TRUE
.
This function allows users to generate predicted probabilities for the
additional outcome variables with the randomized response item as a
covariate given an object of class "rrreg.predictor" from the
rrreg.predictor()
function. Four standard designs are accepted by
this function: mirrored question, forced response, disguised response, and
unrelated question. The design, already specified in the "rrreg.predictor"
object, is then directly inputted into this function.
Blair, Graeme, Kosuke Imai and YangYang Zhou. (2014) "Design and Analysis of the Randomized Response Technique." Working Paper. Available at http://imai.princeton.edu/research/randresp.html.
rrreg.predictor
to conduct multivariate regression
analyses with the randomized response as predictor in order to generate
predicted probabilities.
# NOT RUN { data(nigeria) ## Define design parameters set.seed(44) p < 2/3 # probability of answering honestly in Forced Response Design p1 < 1/6 # probability of forced 'yes' p0 < 1/6 # probability of forced 'no' ## Fit joint model of responses to an outcome regression of joining a civic ## group and the randomized response item of having a militant social connection rr.q1.pred.obj < rrreg.predictor(civic ~ cov.asset.index + cov.married + I(cov.age/10) + I((cov.age/10)^2) + cov.education + cov.female + rr.q1, rr.item = "rr.q1", parstart = FALSE, estconv = TRUE, data = nigeria, verbose = FALSE, optim = TRUE, p = p, p1 = p1, p0 = p0, design = "forcedknown") ## Generate predicted probabilities for the likelihood of joining ## a civic group across respondents using quasiBayesian simulations. rr.q1.rrreg.predictor.pred < predict(rr.q1.pred.obj, avg = TRUE, quasi.bayes = TRUE, n.sims = 10000) # }