Skip to content
Interview prep / ML/Data

Cross-validation for unseen issuers

Question

A model will be deployed on issuers absent from training. Why is a date-only split insufficient to assess that objective?

Short answer

A date-only split can retain the same issuers in training and test, allowing entity-specific relationships to transfer. Hold out relevant issuer groups as well as respecting chronology, and prevent related entities or duplicated records crossing the split. Evaluate future performance on known issuers separately from performance on new issuers. These are different deployment questions and may require different models.

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