Thanks to Kit Baum, an update to the -rrlogit- package is
available from the SSC archive. To install the update,
type
. ssc install rrlogit, replace
-rrlogit- fits logistic regression for randomized response data.
Major changes:
o Individually varying probabilities of unconditional answers
can now be taken into account.
o I accidentally omitted a constant term in the log likelihood
function. This had no implications for the estimated
coefficients, but the reported log likelihood values were
wrong. This is fixed.
o The baseline (constant only) model was estimated incorrectly.
Thus, a wrong value for the overall Chi2-Test was reported.
This is fixed.
ben
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
