Why Automated Validations Fail - podcast episode cover

Why Automated Validations Fail

Jul 05, 202222 minSeason 5Ep. 9
--:--
--:--
Download Metacast podcast app
Listen to this episode in Metacast mobile app
Don't just listen to podcasts. Learn from them with transcripts, summaries, and chapters for every episode. Skim, search, and bookmark insights. Learn more

Episode description

Send us a text

Why does automating model validations fail?

Goals/Plan:
1) Specify the model
2) Runs the development model for a replication check
3) Output of all possible analysis for that model type is produced including challenger models

Reasons it Failed:
1) Data and their processes vary from department to department and change across time without warning.
2) Too much output is generated that needs to be manually reviewed.
3) When issues are found, manual output is still required.
4) Anyone proficient in programming can conduct a full validation faster than the automated system.
5) Manually built models by true experts are still beating automated challenger models or development built models.

Video version:
https://youtu.be/H6kQ7e8whhE



Website:
https://www.FancyQuantNation.com

Quant t-shirts, mugs, and hoodies:
https://www.teespring.com/stores/fancy-quant

Connect with me:
https://www.linkedin.com/in/dimitri-bianco
https://twitter.com/DimitriBianco

Support the show

For the best experience, listen in Metacast app for iOS or Android