Multiple imputation for demographic hazard models with left-censored predictor variables: Application to employment duration and fertility in the EU-SILC
{{output}}
A common problem when using panel data is that an individual's history is incompletely known at the first wave. We show that multiple imputation, the method commonly used for data that are missing due to non-response, may also be used to impute these data that... ...
