Sequential Learning, Asset Allocation, and Bitcoin Returns

Abstract

For optimal asset allocation, mean-variance investors must learn about the joint dynamics of new and existing asset classes, not only their profitability. Bitcoin's digital gold narrative provides a unique laboratory to test this hypothesis. We find that a decrease in investors' estimate on correlation between Bitcoin and the US stock markets strongly predicts higher Bitcoin returns the next day. The same empirical pattern universally appears in out-of-sample predictions, global equity markets, and other cryptocurrencies. Our stylized model and empirical proxy for Bitcoin demand explain the predictability pattern in light of asset allocation practices and investors' learning on time-varying correlation.

Keywords: Bitcoin, Uncertainty, Learning, Time-Varying Correlation, Return Predictability

JEL Classification: G12, G15, D83