Can quanto a quanto risk premium predict BRL/USD? Using the framework of Kremens and Martin (2019) and daily data from February 2015 to March 2026, we construct an empirical proxy from the return differential between unhedged and hedged S&P 500 exposures in BRL and use its smoothed component to forecast exchange-rate movements. In sample, the proxy enters with negative and significant coefficients at the 3–12 month horizons, implying that higher covariance risk predicts subsequent BRL appreciation. Out of sample, re-estimated models fail to beat the random walk, while a theory-imposed forecast delivers positive out-of-sample R2 at the 12 and 24 month horizons. These results show that a quanto risk premium can be recovered from public data and has forecasting value when theoretical restrictions are imposed.
Comissão Organizadora
Comissão Científica
Ferramenta completa de submissão, avaliação e publicação de anais para eventos científicos.
