Bouri, ElieChristou, ChristinaGupta, Rangan2023-11-072023-11-072022-10Bouri E,., Christou, C. & Gupta, R. 2022, 'Forecasting returns of major cryptocurrencies: Evidence from regime-switching factor models', Finance Research Letters, vol. 49, art. 103193, pp. 1-7, doi : 10.1016/j.frl.2022.103193.1544-6123 (print)1544-6131 (online)10.1016/j.frl.2022.103193http://hdl.handle.net/2263/93179DATA AVAILABILITY : Data are available from : https://coinmarketcap.com. The link is mentioned in the manuscript.The returns of cryptocurrencies tend to co-move, with their degree of co-movement being contingent on the (bullish- or bearish-) states. Given this, we use standard factor models and regime-switching factor loadings to forecast the returns of a specific cryptocurrency based on its lagged information and informational contents of 14 other cryptocurrencies, with these 15 together constituting 65% of the market capitalization. Considering top five cryptocurrencies namely, Bitcoin, Ethereum, Ripple, Dogecoin, and Litecoin, we find significant forecastability and evidence that factor models, in general, outperform the benchmark random-walk model, with the regime-switching versions standing out in the majority of the cases.en© 2022 Elsevier Inc. All rights reserved. Notice : this is the author’s version of a work that was submitted for publication in Finance Research Letters. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms are not reflected in this document. A definitive version was subsequently published in Finance Research Letters, vol. 49, art. 103193, pp. 1-7, doi : 10.1016/j.frl.2022.103193.CryptocurrenciesFactor modelsMarkov-switching modelForecastingSDG-08: Decent work and economic growthPreprint Article