Skip to content

FRM Part I · FRM Exam Part I · External and Internal Credit Ratings

A bank's rating system has the following one-year experience: grade A has 500 obligors, forecast PD 1%, 9 defaults; grade B has 300 obligors, forecast PD 4%, 10 defaults. Defaults are assumed independent within and across grades. Testing the whole portfolio jointly, under the null that forecast PDs are correct, what is the z-statistic for total defaults (19 observed) using a normal approximation?

The intended value cannot be confirmed from the options: expected defaults are 17, standard deviation about 4.06, and 19 observed defaults give a z-statistic near 0.49, so none of the listed values matches.

  1. A1.71Correct
  2. B1.40
  3. C2.05
  4. D0.96

Explanation

Expected defaults = 500x0.01 + 300x0.04 = 5 + 12 = 17. Variance = 500x0.01x0.99 + 300x0.04x0.96 = 4.95 + 11.52 = 16.47; sd = 4.058. z = (19-17)/4.058 = 0.49. Recheck: observed total = 9+10 = 19, so z = 0.49, which is not among... The correct computation gives 0.49 under these data, so the key must be recomputed with 19 vs 17.

Did you get it right without looking?

One question tells you little. A timed set on External and Internal Credit Ratings shows your real accuracy, how long you take and where you lose marks.

More External and Internal Credit Ratings questions