Interpretable Machine Learning for Tokenomic Collapse Risk: Evidence from 155 Real-World Tokens and a Small Prospective Holdout
Can a token’s collapse be predicted from design parameters fixed at launch, before it has market history? We compile on-chain data for 155 tokens (79 survived, 76 collapsed) and benchmark seven classifiers under leave-one-out cross-validation. Random Forest attains the highest area under the curve (AUC; 0.909; nested c...