Versioned data on Hugging Face; compact fixtures on GitHub.
The complex targeting model has not earned its advantage.
I test effect estimates against known answers, then compare spending rules at the same budget in simulated markets held out from training.
Read the study & evidence →Most factors do not travel; funding and positioning hold up better.
I test cross-sectional factors on 105 coins, check split samples and parameters, compare funding on OKX, and track the composite after publication.
Read the study & evidence →Pooling explains much of the apparent predictability.
I compare out-of-sample linear, tree, and neural-network forecasts on JKP global factors, test developed-to-emerging-market transfer, and evaluate timing after turnover costs.
Read the study & evidence →Changing the data vintage changes which models look best.
I reconstruct official release histories and rerun forecasts with three information rules: data known then, revised values with the same availability, and unrestricted latest values.
Read the study & evidence →Order flow carries a positive gross edge in 110 of 144 execution scenarios.
I replay one four-hour BTC/ETH order-book capture across two evaluation phases, four horizons, and nine delay combinations, with a fixed 4 bp fee.
Read the study & evidence →Reversal, volatility, and illiquidity survive the corrections.
I study 4,000 Polymarket markets, remove five backtest artifacts, control for price levels, and evaluate factor signs on a held-out half of the panel.
Read the study & evidence →Bigger rate gaps are associated with fewer mortgage exits.
I compare monthly payoff rates across rate gaps, then estimate a complementary log-log hazard model with observed loan and local-market controls.
Read the study & evidence →The evidence points toward higher costs for U.S. importers.
I compare each wave of tariffed products with never-tariffed products, tracking unit import values before and after the policy and checking three sample windows.
Read the study & evidence →