[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Population Attributable Fraction (PAF) estimation
Dear Statalist members,
I have encountered considerable problems in using the 'aflogit.ado'
program (STB-42 sbe21) to estimate adjusted PAFs . These are
some of them:
1) Unable to estimate PAFs following conditional logistic
2) Unable to estimate PAFs following poisson regression with
aggregated person-years data and interaction terms between
3) Error message ("option cc not allowed") when trying to use the
cc option (i.e. "aflogit, cc") to specify that data are from a case-
control design rather than a cross-sectional or cohort study.
4) Generation of invalid estimates when the 'if' command is used to
restrict the data set
Does anyone know a of simple and more reliable way of obtaining
adjusted PAFs following logistic, poisson and Cox regression
Are there any plans to incorporate a flexible and valid adjusted PAF
estimator into future versions of Stata that can deal with a variety of
different regression models?
Is this currently possible in any other statistical packages?
Thanks and best wishes,
University of Manchester (UK)
* For searches and help try: