Financial volatility modeling and forecasting has always been a hot topic in financial econometrics， due to its great importance for derivative pricing， asset allocation and risk management. Typically， GARCH model is used to describe the dynamics of financial volatility. However， the GARCH model uses squared return to measure volatility， ignoring the information of intraday price movements. An alternative approach for measuring volatility is to employ the intraday range， which is calculated using the intraday high and low prices. Apparently， the intraday range makes full use of the intraday price information （extreme value information）， which is a more efficient volatility estimator than the squared return volatility estimator.A classical model for describing the dynamics of the intraday range is the conditional autoregressive range （CARR） model， which produces more accurate volatility forecasts than the return-based GARCH model. Despite the empirical success of the range-based CARR model， it cannot capture the impact of macroeconomic variables （macroeconomic information） on financial volatility. In recent years， the level of economic policy uncertainty （EPU） keeps rising， due to a series of events including the US-China trade war and the coronavirus （COVID-19） pandemic. Intuitively， high EPU may affect investors' investment decisions and hence financial market. The foreign exchange market is one of the largest and most liquid financial markets in the world， which is of great relevance for investors and policy-makers and would have a close relation to EPU. As the currency of the world's second largest economy， renminbi plays a more and more important role in the world economy. Since the implementation of renminbi exchange rate regime reform in 2005， the renminbi exchange rate has experienced significant fluctuations. Accurate prediction of the renminbi exchange rate volatility has become increasingly important. To our knowledge， there are few studies investigating the impact of EPU on the renminbi exchange rate volatility.Inspired by the return-based GARCH-MIDAS model， this paper extends the classical range-based CARR model to the range-based CARR-MIDAS model to model the renminbi exchange rate volatility. The model framework explores the intraday extreme value information and allows the low-frequency macroeconomic variable （macroeconomic information） such as EPU directly impacts the volatility via the long-run component of volatility and the flexible MIDAS structure.Using the monthly global EPU index and daily US Dollar against Chinese Yuan （USD/CNY） exchange rate data， the impact and predictive ability of the EPU on USD/CNY exchange rate volatility are investigated relying on the range-based CARR-MIDAS model with the EPU （CARR-MIDAS-EPU）. The empirical results show that the EPU has a significant positive impact on the long-run volatility of USD/CNY exchange rate. That is， an increase in the EPU level predicts higher level of the long-run volatility of USD/CNY exchange rate. The range-based CARR-MIDAS-EPU model produces more accurate out-of-sample forecasts of the USD/CNY exchange rate volatility compared to a variety of competing models， including the return-based GARCH model， GARCH-MIDAS model and GARCH-MIDAS-EPU model as well as the range-based CARR model and CARR-MIDAS model， for forecast horizons of 1 day up to 3 months. This finding suggests that the range and EPU contain valuable information for forecasting USD/CNY exchange rate volatility. The robustness analysis based on the alternative global EPU index as well as the out-of-sample forecasting windows further supports the above conclusion.
