Title: Small sample proportional hazards inference, with application to mortgage prepayment
Authors: John Kolassa - Rutgers, the State University of New Jersey (United States) [presenting]
Abstract: Proportional hazards regression techniques are often used to model event time data subject to censoring. Small samples involving discrete covariates with strong effects can lead to infinite maximum partial likelihood estimates. A methodology is presented for eliminating nuisance parameters estimated at infinity using approximate conditional inference. Conventional higher-order likelihood inference may then be applied to remaining parameter components. Techniques will be applied to models for mortgage prepayment.