Sep 10

Altman’s Z score – 8 min screencast

by David Harper, CFA, FRM, CIPM

Altman’s Z is the most famous type of linear discriminant model: borrowers are classified into high or low default risk categories. It does not directly give a probability of default (PD), although we can map to the score to a credit rating and map the rating to a PD (so there is an indirect path from the score to the PD).

Saunders gives four drawbacks, though they implicate discriminant models generally more than Altman’s specifically:

Not granular: only gives default/zone of ignorance/no default
Constant factor weights (i.e., factor weights may be time varying)
Only considers five fundamental variables, ignores other variables
No centralized database on defaulted business loans (not really an Altman’s critique at all)

Screencast: