Skip to content
Interview prep / ML/Data

Calibration at the traded threshold

Question

A classifier is well calibrated overall but trades only when predicted probability exceeds 0.8. What should be checked?

Short answer

Inspect calibration, sample size and payoff distributions specifically near and above the trading threshold. Global calibration can hide errors in the small traded subset or in particular regimes. Evaluate after any probability recalibration and threshold selection on independent data. Even accurate 80% event probabilities do not establish profitability without event magnitudes, costs and a feasible sizing rule.

Original question and answer written by SelfMadeQuant.com.

See the hint and the full worked solution in the question bank (free account).

Open the worked solution

Lessons behind this question

Start with a free lesson: Expectation, Variance and Covariance is part of a free topic. Create a free account, no card needed, to read it in full with worked solutions.

Read it free