Climate Risks and Predictability of the Conditional Distributions of Rare Earth Stock Returns and Volatility

We use a k-th order nonparametric causality-in-quantiles test, we predict rare earth stock returns and volatility due tobased on physical and transition climate risks over the period of 2nd January 2008 to 31st January 2025. The results indicate that, even though linear Granger causality fail to show any evidence of prediction of rare earth stock returns, due to model misspecifications from nonlinearity and structural breaks, the nonparametric framework depicts statistically show significant evidence of predictability over the entire conditional distribution of returns and volatility. They are robust to alternative choices of rare earth stock indexes, measures of climate risks, conditional estimates of volatility, and various macroeconomic and financial control variables. Further analyses involving the signs of the causality and rolling-window estimation tend to reveal that rare earth stock index returns are negatively impacted at lower conditional quantiles till the median, corresponding to bearish market conditions; volatility, however, is positively impacted (i.e., it increases) over its entire conditional distribution.