PaMIR benchmark released for credit-default prediction models

Researchers have released PaMIR, an open benchmark designed to evaluate credit-default prediction models, particularly in scenarios with limited and delayed label data. This benchmark consolidates 19 public datasets, encompassing over 1.24 million loan and account records from nine countries, all processed using a single, audited methodology. PaMIR evaluates models based on their performance in a simulated streaming environment where each application is scored upon arrival, reporting AUC by label budget. AI

IMPACT Provides a standardized evaluation framework for credit-default prediction models, potentially improving their reliability and adoption.

RANK_REASON The item describes the release of a new benchmark dataset and methodology for machine learning research. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• CORE Recommender
• DagsHub
• Gotit.pub
• Hugging Face
• IArxiv Recommender
• Influence Flower
• PaMIR
• ScienceCast
• Shuhratjon Khalilbekov

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →