Original scientific paper

PREDICTING BANKRUPTCY WITH SEMI-PARAMETRIC SINGLE-INDEX MODEL

Arjana Brezigar Masten
Igor Masten

Abstract

Semi-parametric methods are virtually neglected in the bankruptcy prediction literature. This paper compares the logit model, as the standard parametric model for bankruptcy prediction, to the semi-parametric model developed by Klein and Spady (1993). Special care is devoted to the effect of choice-based sampling prediction accuracy. The choice of the sampling and estimation method lead to a similar trade offs. Using choice-based sampling and logit model leads to minimization of risk exposure. Samples unbalanced across groups and the semi-parametric method allow for better overall prediction accuracy and thus profit maximization.

Keywords

bankruptcy prediction; semi-parametric methods

Hrčak ID:

80016

URI

https://hrcak.srce.hr/80016

Publication date:

1.4.2012.

Article data in other languages: croatian

Visits: 2.184 *