In order to compute first differences and risk ratios, you need two "setx"
objects. For example:
x.out <- setx(MH6)
x1.out <- setx(MH6, L_PC_PRODUCTION = 1)
sim(MH6, x = x.out, x1 = x1.out)
I'm not sure of the status of the plotting functions for relogit. The plot you are
looking for simply may not be implemented.
However, still try:
plot(MH6)
On Dec 5, 2011, at 2:13 PM, <alero(a)utdallas.edu> <alero(a)utdallas.edu> wrote:
Hello,
After running my Zelig relogit model and simulating values from the posterior, I am
interested in getting predicted probabilities, first differences, and risk ratios. From my
simulation results, I am only able to get expected values and predicted probabilities of
my covariates. Am I not able to get risk ratios and first differences from the relogit
model? Also, for the plot command, am I not able to get plots of individual posterior
densities for each covariate? The plot function gives me just one graph for the Expected
Values: E(Y|X). See code below:
MH6 <- zelig(onset ~ L_PC_PRODUCTION + L_PC_RESERVES + L_PC_DIAMONDS
+ L_wdiagdp, data = mhd, model = "relogit", tau = 88/5269,
case.control = "weighting", robust = TRUE)
summary(MH6)
# Setting explanatory variables to mean
x.MH6 <- setx(MH6)
# Simulating data
s.MH6 <- sim(MH6, x = x.MH6, num = 100000)
summary(s.MH6)
plot(s.MH6)
Thank you,
Alero