摘要：Many researchers see the need for reject inference to come from a sample selection problem whereby a missing variable results in omitted variable bias. Specifically, the success in being accepted for a loan is related to subsequent repayment performance. Accordingly, the residuals of the previous scoring model by which the person is accepted may be correlated with those of a new model that predicts his repayment performance. Unless the correlation between the residuals of the new and old model are reflected in the new model its parameters will be biased. Alternatively, practitioners often see the problem as one of missing data where the relationship in the new model is biased because the behaviour of the omitted cases differs from that of those who make up the sample for a new model.
Loading...
