It dawned on me that the illustrative do-file appended to the earlier post
omitted a line of code.
The lines
quietly centile xg, centile(5)
quietly replace a = .b if a < r(c_1)
label define LOQ .b "< LOQ"
label values a LOQ
intcens a b, distribution(gam) nolog
need to include one setting the right-hand response variable to the limit of
quantitation (detection) in order for -intcens- to give correct estimates
quietly centile xg, centile(5)
quietly replace a = .b if a < r(c_1)
quietly replace b = r(c_1) if b < r(c_1) // <- here
label define LOQ .b "< LOQ"
label values a LOQ
intcens a b, distribution(gam) nolog
So, for values below the hormone assay's minimum detectable value (MDL),
Leny would set a as missing and b to the MDL, followed by the predictors.
Joseph Coveney
*
* 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/
