Reproducibility Package for “Does Multi-Signal Cryptocurrency Screening Add Value? Point-in-Time Out-of-Sample Evidence against Parsimonious Risk Rules”
Description
This dataset is the reproducibility package accompanying the study “Does Multi-Signal Cryptocurrency Screening Add Value? Point-in-Time Out-of-Sample Evidence against Parsimonious Risk Rules.” The study evaluates whether a seven-signal cryptocurrency screening mechanism provides incremental out-of-sample value beyond parsimonious risk-based selection rules. The empirical design reconstructs 48 monthly point-in-time investable universes from July 2022 through June 2026. At each formation date, eligible cryptocurrency assets are identified from dated historical market snapshots and ranked using liquidity, 30- and 90-day returns, realised volatility, downside deviation, maximum drawdown, and historical 95% conditional value-at-risk. The complex screening mechanism combines monthly MEREC weighting with CoCoSo aggregation and is evaluated against transparent risk, momentum, equal-weight, investable-universe, and Bitcoin benchmarks. The reproducibility package contains the analysis code, derived dated snapshots, daily market panel, selection and eligibility audit information, decision matrices, rankings, portfolio and strategy returns, statistical-test outputs, robustness results, and figures used in the study. The materials support reproduction and auditing of the reported empirical results. Original source data obtained from CoinMarketCap, Binance, and FRED remain subject to the respective providers’ terms of use. The deposited materials therefore consist of code and derived research data included in the reproducibility package rather than a redistribution claim over third-party source data.
Files
Institutions
- Nevşehir Hacı Bektaş Veli UniversityNevşehir Province, Nevşehir