Several studies suggested that there is a significant relationship between uncertainty indexes and commodities, and reported that this relationship varies across different types of commodities and uncertainty measures. This highlights the importance of considering uncertainty when analyzing commodity prices and making investment decisions in this market. For instance, Assaf et al. (2021) investigated the relationship between the energy market and uncertainty indexes using the time-domain TVP-VAR-based connectedness approach. Their main focus is to determine whether four distinct types of uncertainty including Economic Policy Uncertainty (EPU), Geopolitical Risk (GPR), World Trade Uncertainty (WTU), and Equity Market Uncertainty (EMU) have an effect on the returns of the oil, gas, and coal markets. They found that market uncertainty has an average impact of approximately 53% on energy market and EPU has the most significant influence on it followed closely by the World Trade Uncertainty index. Similarly, Lyu et al. (2022) employed a time-varying parameter vector autoregressive (TVP-VAR) model to examine how economic policy uncertainty shocks impact China's commodity futures returns. They confirmed that the reactions of different sectoral commodity futures returns to EPU shocks are significant and vary over time. In the same way, Yang et al. (2022) examined the response of commodity market prices to trade policy uncertainty and geopolitical shocks and found that both variables have significant time-varying effects on the aggregate commodity market. Similarly Huang et al. (2021) used a time-varying parameter vector autoregression (TVP-VAR) model to examine the dynamic relationships between commodity prices and different measures of uncertainty, including economic policy uncertainty, macroeconomic uncertainty (MU), equity market uncertainty (EMU), and stock market volatility (VXO). The study found that while the effects of uncertainty shocks on commodity prices are time-varying for most measures, MU does not exhibit significant time variation. Zhou et al. (2023) employed also a time-varying parameter vector autoregressive model with stochastic volatility (TVP-SV-VAR) to investigate the dynamic relationship between climate policy uncertainty, oil prices, and renewable energy consumption. The study revealed a time-varying relationship between the variables considered.
On the other hand, other studies employed different research methods. For instance Liu et al. (2022) divided uncertainty into economic, geopolitical, market, natural disaster, and public health dimensions and examined how the international commodity market responds to those different types of uncertainty. Using a wavelet analysis, they found that the international commodity market is more substantially affected by uncertainty over shorter periods. Besides, using a Rolling Tail-Event driven NETwork, Ren et al. (2022) identified an asymmetric impact of economic policy uncertainty on the risk spillover of various commodity futures markets. The degree of this impact varies across different quantiles. Quin et al. (2020) examined the asymmetric effects of geopolitical risks on the returns and volatility of energy commodities, including crude oil, gas, and heating oil by employing a quantile regression model. Their findings suggested that geopolitical risks have significantly negative effects on crude oil returns in the bearish market and on heating oil returns in the normal and bullish markets. However, there were no significant effects observed on gas returns. Furthermore, the study highlighted that geopolitical risks have a significantly positive impact on crude oil volatility across different market conditions. The impacts on gas and heating oil volatility are mainly negative but not statistically significant.
While the above-mentioned researches investigated the commodity markets as a whole, there are also studies in the literature focusing on precious metal markets and their interaction with uncertainty measures. For instance, Bilgin et al. (2018) employed a nonlinear Autoregressive-distributed Lag (ARDL) model and found that gold prices respond positively to economic policy uncertainty. Similarly, Bouoiyour et al. (2018) found a positive and strong relationship between gold returns and different uncertainty measures including economic, macroeconomic, microeconomic, monetary policy, financial and political uncertainties which is not consistent with the findings of Reboredo and Uddin (2016), who showed that there is no co-movement, and no Granger causality between precious metals commodity futures, policy uncertainty and financial uncertainty measured by the VIX.
However, the results of the study conducted by Chai et al. (2019) reported that the effects of global economic policy uncertainty shocks on gold prices change over time, suggesting that the reliability of gold as a safe haven is not constant. In their study, Yilanci and Kilci (2021) examined the causal relationship between economic policy uncertainty, geopolitical risk, and precious metal prices. Their findings suggested that the causality links from geopolitical risk and economic policy uncertainty to the prices of precious metals vary over the analysis period.
Chiang (2022) used a GED-GARCCH model to determine whether gold could be used as a hedge against different forms of uncertainty, including economic policy uncertainty, geopolitical risks, and equity market volatility. The results indicated that gold can serve as a hedge against uncertainty, as it exhibited a positive response to an increase in the set of uncertainty variables. Huynh's (2020) study stands out in the literature for utilizing two advanced methodologies, namely Multilayer Perceptron Neural Network Non-linear Granger causality and Transfer Entropy, to investigate the causal relationship between Economic Policy Uncertainty (EPU), the Chicago Board Options Exchange Volatility Index (VIX), and precious metals prices such as gold, silver, palladium, and platinum. The study discovered that gold remains the leading "safe-haven" asset for hedging against uncertainty. Notably, the precious metals analyzed were found to influence EPU and VIX, with immunity to EPU shocks but not to those originating from the VIX. According to Huang et al. (2022), precious metal returns were positively impacted by oil price shocks before the international financial crisis, but these effects have turned negative since then. The influence of economic policy uncertainty on precious metal returns is subject to temporal fluctuations, tending to be positive in the majority of cases. Furthermore, the study found that economic policy uncertainty can amplify the effects of oil price shocks on precious metal returns. Baur and Smales (2020) found that holding precious metals within a diversified portfolio lowers the impact of geopolitical risk.
In addition, some studies have sought to investigate how the COVID-19 period has shaped the relationship between commodities and measures of uncertainty. For instance, Mokni et al. (2021) employed a Quantile VAR model to investigate how economic policy affected the interdependence of precious metals including gold, silver, palladium, and platinum during the COVID-19 pandemic. The authors discovered that gold remained the primary safe-haven asset for hedging market uncertainty, while silver, platinum, and palladium exhibited varied responses to the pandemic. Raza et al. (2023) found a significant correlation between global GEPU and precious metal price volatility before and during the COVID-19 period, using a GARCH-MIDAS approach.
At this point, it should be noted that limited studies have examined the relationship between commodities and the cryptocurrency market. Hassan et al. (2021) employed a Dynamic Conditional Correlation Generalized Autoregressive Conditional Heteroskedasticity model to examine the hedging properties of precious metals against cryptocurrency uncertainty indexes. They concluded that gold remains the leading safe-haven asset for hedging against uncertainty. Using a TVP regression-based Dynamic Occam's Window (DOW) forecasting framework, Shang et al. (2022) stated that cryptocurrency policy uncertainty helps to predict weekly gold returns. Elsayed et al. (2022) assessed the connectedness between gold, cryptocurrency uncertainty indices, economic and market uncertainty indices using the Time-Varying Parameter Vector Autoregressive. They found that cryptocurrency policy uncertainty is the major transmitter of return spillovers. They reported also that gold fails to hedge the uncertainty shocks from cryptocurrencies. Recently, Wei et al. (2023) examined whether cryptocurrency policy and price uncertainty can help to forecast volatilities in precious metal markets. Using a GARCH-MIDAS model, they found that these uncertainty measures have a significant impact on the volatilities of gold and silver and are potentially better predictors of daily volatility in precious metal markets compared to other commonly used uncertainty indexes. Since the above mentioned studies highlighted a significant connection between cryptocurrency uncertainty indexes and commodities, we hypothesize the following:
H1. Cryptocurrency uncertainty indexes are significant predictors of white precious metal returns.
H2. White precious metal returns respond to shocks from cryptocurrency market.