Search academic texts

Information × Registration Number 2123U006682, Article popup.category Препринт Title popup.author Kyba Sofia popup.publication 01-01-2023 popup.source_user Український католицький університет popup.source https://hdl.handle.net/20.500.14570/4810 popup.publisher Description We analyze the relationship between cryptocurrency returns and common risk fac- tors by implementing Instrumented Principal Component Analysis model, which stands for latent factor model with dynamic time-varying loadings. We build upon Kelly, Pruitt, and Su, 2019, compare a new approach with traditional observable factor models, and show that four-factor IPCA indeed appears to outperform the ordinary approach. The IPCA model with four latent factors explains around 14% (46%) of the variation in the returns of individual cryptocurrencies (manage portfo- lios) compared to the less than 10% (15%) explained by the observable factor models. The IPCA factors proxy the market return, lottery demand, liquidity, volatility, and downside risks. Link to the github repository. popup.nrat_date 2025-05-09 Close