Dear all,
I am using a dataset in which only about 600 individuals out of 25000
answer "yes" to a specific question. I have not found any warning in the
literature against the use of the usual probit or logit modelling when
there are much less outcomes ("ones") than non-outcomes ("zeros"), or any
alternative to those models that would be more appropriate to this
particular problem. I would appreciate very much any comments on this.
Best regards,
Christine Valente
Doctoral candidate
Department of Economics
University of Sheffield
9, Mappin Street
Sheffield S1 4DT
U.K.
+44(0)7976866835 (mobile)
+44(0)1142223321 (office)
|