Time-frequency connectedness of crude oil,economic policy uncertainty and Chinese commodity markets: Evidence from rolling window analysis

This article investigates the time–frequency connectedness of economic policy uncertainty (EPU), WTI crude oil and Chinese commodity markets during the period between 2004 and 2020. Rolling window wavelet vector autoregression and connectedness networks are developed to evaluate the time-varying characteristics of the connectedness. The empirical results are as follows: First, the total connectedness between EPU, oil and commodities becomes stronger as the time scale increases. Second, the net connectedness of EPU and WTI in the system is positive, indicating that EPU and WTI are contributors to information and will affect financial markets across time scales. Third, the connectedness remains at a high level during financial crises across all scales, and the contribution of EPU and crude oil to commodities increases significantly. Specifically, compared with other commodity sectors, grains are greatly affected by EPU under the condition that the energy sector is seriously affected by crude oil. Overall, investors and policy makers should consider connectedness in terms of time and frequency when making a decision.     

Jump Interdependencies: Stochastic linkages among international stock markets

This study investigates the volatily jump contagion among the Asian, European (Germany, UK, & France) and US markets. In particular, it examines the stochastic linkages among the international stock markets and analyzes the self and cross-excitation of jumps. The discontinuities in the stochastic volatility of each market are identified and their structural inter-dependencies are analyzed. Our empirical results imply that negative jumps from the USA and Europe are transmitted to the domestic Asian markets, while positive jumps are majorly from the regional markets. Results also imply that the cross-market linkages vary with respect to markets and regimes. Our results have implications for risk management, investment and hedging decisions.     

Multiscale stock-bond correlation: Implications for risk management

This paper examines the multiscale return correlation between the stocks and government bonds of different maturities returns in 25 countries. The analysis reveals that developed markets correlations are generally negative at the first time-scale and move in a positive direction at higher scales. This contrasts with emerging markets, where the correlation tends to be positive throughout. Thus, the results support a greater flight-to-safety effect in developed markets. Further evidence highlights the ability of the correlation to produce portfolios with a lower VaR. Results support this at longer time-scales and for both developed and emerging markets. The results here demonstrate the importance of accounting for time-scales in modelling the stock-bond correlation and in constructing portfolios.     

Can crude oil drive the co-movement in the international stock market? Evidence from partial wavelet coherence analysis

In this paper, we employ partial- and multiple-wavelet coherence analyses to examine co-movement between international stock markets by considering the influence of crude oil in a time domain perspective. Overall, we find that crude oil is a major factor driving co-movement between international stock markets in the median and long term. However, when considering the oil-importing and oil-exporting countries differently, we still find that crude oil is a driver for interdependence between oil-importing and oil-exporting countries. In contrast, the crude oil has relative lower impact on the co-movement in oil-importing or in oil-exporting countries, which indicates its co-movement is caused by other factors. In addition, Gulf Cooperation Council stock market may lead the stock markets of oil-importing countries in the long term. Our empirical results provide meaningful information for investors and policymakers.     

Volatility cascades in cryptocurrency trading

This paper studies volatility cascades across multiple trading horizons in cryptocurrency markets. Using one-minute data on Bitcoin, Ethereum and Ripple against the US dollar, we implement the wavelet Hidden Markov Tree model. This model allows us to estimate the transition probability of high or low volatility at one time scale (horizon) propagating to high or low volatility at the next time scale. We find that when moving from long to short horizons, volatility cascades tend to be symmetric: low volatility at long horizons is likely to be followed by low volatility at short horizons, and high volatility is likely to be followed by high volatility. In contrast, when moving from short to long horizons, volatility cascades are strongly asymmetric: high volatility at short horizons is now likely to be followed by low volatility at long horizons. These results are robust across time periods and cryptocurrencies.     

Dependence structures between Chinese stock markets and the international financial market: Evidence from a wavelet-based quantile regression approach

In this study, we investigate the dependence structures between six Chinese stock markets and the international financial market including possible safe haven assets and global economic factors under different market conditions and investment horizons. The research is conducted by combining a quantile regression approach with a wavelet decomposition analysis. Although we find little or insignificant dependence under short investment horizons, we detect the strong asymmetric dependence of oil prices and the US dollar index on the six Chinese stock markets in the medium and long terms. Moreover, not only is crude oil not a safe haven, it may damage Chinese stock markets as it increases over the long term, even in bull markets. Meanwhile, appreciation of the US dollar (depreciation of RMB) damages (boosts) Chinese stock markets during bull (bear) market conditions under long investment horizons. Moreover, we find that VIX (volatility index)-related derivatives may serve as good risk management tools under any market condition, while gold is a safe haven asset only during crisis periods.     

A neural network enhanced volatility component model

Volatility prediction, a central issue in financial econometrics, attracts increasing attention in the data science literature as advances in computational methods enable us to develop models with great forecasting precision. In this paper, we draw upon both strands of the literature and develop a novel two-component volatility model. The realized volatility is decomposed by a nonparametric filter into long- and short-run components, which are modeled by an artificial neural network and an ARMA process, respectively. We use intraday data on four major exchange rates and a Chinese stock index to construct daily realized volatility and perform out-of-sample evaluation of volatility forecasts generated by our model and well-established alternatives. Empirical results show that our model outperforms alternative models across all statistical metrics and over different forecasting horizons. Furthermore, volatility forecasts from our model offer economic gain to a mean-variance utility investor with higher portfolio returns and Sharpe ratio.     

Co-movement between the US and the securitised real estate markets of the Asian-Pacific economies

The novelty of this study is the use of continuous wavelet transform analysis of wavelet coherence, as well as its partial and multiple forms, to revisit the co-movements of Asian-Pacific public real estate markets among themselves and with the US, for a time span which covers the 12 January 1995–23 June 2016 period. Earlier research does not have satisfactory results because traditional methods average different relationships in time domain only. From the wavelet analysis, investors can extract the time-scale that most interests them. We find that the co-movement relationship across the real estate markets increases during the two major crisis period, as well as becomes stronger as the scale increases. Hong Kong and Singapore have the strongest time-scale co-movement relationship. Finally, the influence of domestic macroeconomic factors on real estate return co-movement appears to be greater at the long-term horizons than at the short-term horizons.     

Market volatility and illiquidity during the COVID-19 outbreak: Evidence from the Saudi stock exchange through the wavelet coherence approaches

The aim of this paper is to examine the explanatory power of realized volatility on the illiquidity in Saudi stock market during the COVID-19 outbreak. To achieve this objective, we consider the Wavelet Coherence approaches as empirical tools to investigate the combined effect of realized volatility and COVID-19 counts on the market illiquidity across frequencies and over time space by taking in account the number of infected cases in Saudi Arabia and over the World, and the number of death cases in Saudi Arabia as well as over the World. Our study reaches two main findings. First, the preliminary results reported by the ARDL bound test as a benchmark model showed significant long-run and short-run effects of the market volatility on illiquidity in contemporaneous and lagged manner. Second, the wavelet coherence analysis tools exhibited important results: (i) the wavelet coherency between illiquidity ratio and realized volatility in Saudi Arabia appear highly pronounced over all time horizons. (ii) PWC plots showed a significant mutual effect between liquidity risk and realized volatility when eliminating the effect of local COVID-19 cases. (iii) MWC plots highlighted that the response of the market illiquidity index to both the amplification in confirmed local cases (resp. international confirmed cases) and the stock market volatility appear significant in the short and middle horizons.     

A comparative study of exchange rates and order flow based on wavelet transform coherence and cross wavelet transform

Understanding the relationship and behavior of microstructures and exchange rates is an essential discussion for global foreign exchange investors. There are numerous research works regarding the linkage between order flow and exchange rates, yet the exact relationship between the order flow and the market price that indicates whether participants have an impact on the market trend remains undefined. This paper investigates the empirical association and behavior of order flow and the exchange rate movements within the time-frequency space, on three popular currency pairs, using the cross wavelet transform and wavelet transform coherency. The results indicate that order flow has a strong negative correlation and is the leader variable of the exchange rate. A predictor using order flow as an input variable was implemented to forecast the exchange rate direction using the sample data and out-of-sample data. This methodology, which performs with high accuracy and very low drawdown, could be a suitable tool for portfolio managers and forex participants during their trading activities.