What's new

z-score calibration (adjustment)

Thread starter #1
Hi @David Harper CFA FRM ,

One formula I am struggling to understand is the adjustment to the z-score to account for the costs involved with the type I and type II errors ( => opportunity cost vs. LGD) in De Laurentis - Ch3 (Ratings Assignment Methodologies) pp 59 and 60.
ln(q(solvent) * opportunity cost / q(insolvent)* LGD) vs. ln(q(solvent)/q(insolvent)) if we don't want to take into account the costs.

If LGD > opportunity costs, the formula decreases the adjustment, making it more likely to lend to firms that would default afterwards (compared to an adjustment calculated without incoporating the costs)... One would expect the opposite, since a logic response to a higher LGD than opp costs is trying to reduce this risk even if it means assuming more opportunity costs.

I don't know how crucial is this part of the curriculum but it would be good to clarify this. So, I would be grateful if you can help :)


David Harper CFA FRM

David Harper CFA FRM
Staff member
Hi @uness_o7 Apologies but I haven't examined carefully his Cost of Classification; it's not technically in the LO syllabus, right? We didn't include in our notes .... That looks interesting, I will take a closer look what I get a chance ... thanks,
I have the same problem. Adjustment for classification errors should be the inverse of what’s indicated in the readings. Hope there won’t be a question about that as you never know whether they require the “correct” answer or the “curriculum” answer.