Rolling Window Analysis Research Articles

Dynamic Cross-Correlations Analysis on Economic Policy Uncertainty and US Dollar Exchange Rate: AMF-DCCA Perspective

In this paper, we employ the multifractal detrended cross-correlation analysis (MF-DCCA) as the measurement instrument for the dynamic cross-correlation inspection between US economic policy uncertainty (EPU) index and US dollar exchange rate return (Ret). By calculating the cross-correlation statistics, we find mild acceptance of cross-correlation between EPU and Ret qualitatively. With further application of MF-DCCA methodology, we find strong power law cross-correlation existence within all scaling orders. Also, apparent persistence of cross-correlation has been discovered with significant Hurst exponents of all orders. Besides, we find that long-term cross-correlation demonstrates more persistence and higher degree of multifractality than those in the short term. Finally, we utilize the rolling window and binominal measurement analysis as revisits of the model. The results are consistent with model statements.

Dynamic Cross-Correlation between Online Sentiment and Stock Market Performance: A Global View

This paper focuses on investigating the dynamic cross-correlation relationship between online sentiment and returns of major global stock markets based on the MF-DCCA method. We use Daily Happiness (DHS), an index derived from Twitter posts through textual analysis as a proxy of online sentiment. By dividing the global financial markets into developed and developing ones, we are able to test the heterogeneous relationship between stock market performance and sentiment at different economic developing level. Empirical results show that there exists a power-law cross-correlation relationship between financial market and online sentiment in some developed countries and all developing countries, and the relationship is more stable in the developing countries. Moreover, we apply rolling window analysis to capture the dynamic evolution characteristics and find the relationship has a strong consistency over time. Our work provides a much more delicate perspective to test the relationship between online sentiment and financial markets performance and enriches the existing literature.

Research on dynamic structure of the exchange rate volatility network among the Belt and Road countries based on spillover effect

ABSTRACT The countries taking part in China’s ‘The Belt and Road’ Initiative have achieved greatly increased trade volume and become more firmly interconnected since the initiative was launched in September 2013. This paper researches the exchange rate volatility features among 15 countries along the Belt and Road region with the network analysis based on spillover effect. A directed network is constructed to show not only the impact one currency having on and received from the other currencies, but also the net effect from one currency to another in the network. The importance of any currency in the network and the total interconnectedness of the whole network can also be obtained. Furthermore, a 200-day rolling window analysis is used to show the dynamic evolution trend of the network. The results show that after the launch of the Belt and Road Initiative, both the interconnectedness of the exchange rate volatility network and the importance of CNY increase. In addition, the importance of most currencies in the network has been quite volatile.

Can fiat currencies really hedge Bitcoin? Evidence from dynamic short-term perspective

Whereas much research has largely investigated the safe haven, diversifier and hedge proprieties of cryptocurrency, very few papers have analyzed the hedging issue of cryptocurrency with other assets. As such, this paper attempts to investigate the possibility if Bitcoin can be hedged by selected fiat currencies (EUR, JPY and GBP) as Bitcoin prices have experienced high and persistent volatility. To do so, we compute optimal hedge ratios between Bitcoin and fiat currencies over the period 02/02/2012–30/11/2017 based on the VAR-DCC-GARCH model, VAR-ADCC-GARCH model and VAR-component GARCH-DCC model. A rolling window analysis is employed to establish out-of-sample one-step-ahead forecasts of dynamic conditional correlations between different assets. This leads to establish time-varying hedge ratios and thus dynamic cross-hedging Bitcoin/fiat currency markets. The empirical results clearly show the time-varying correlations between Bitcoin and fiat currencies under different specifications, implying a dynamic behavior of the relationship between such assets. For all the proposed models, such dynamic correlations are rather characterized by trending downward over the period under study. The results also display time-varying hedge ratios which lead to an ongoing regular demand for rebalancing the hedged positions under different specifications. As a matter of fact, using various models which take into account different aspects of volatility and correlation structures allows to better implement dynamic hedging strategies.

Analyzing the Impact of COVID-19 on the Cross-Correlations between Financial Search Engine Data and Movie Box Office

This paper explores the COVID-19 influences on the cross-correlation between the movie market and the financial market. The nonlinear cross-correlations between movie box office data and Google search volumes of financial terms such as Dow Jones Industrial Average (DJIA), NASDAQ and PMI are investigated based on multifractal detrended cross-correlation analysis (MF-DCCA). The empirical results show there are nonlinear cross-correlations between movie market and financial market. Metrics such as Hurst exponents, singular exponents and multifractal spectrum demonstrate that the cross-correlation between movie market and financial market is persistent, and the cross-correlation in long term is more stable than that in short term. In the COVID-19 period, the multifractal features of cross-correlation become stronger implying that COVID-19 enhanced the complexity between the movie industry and the financial market. Furthermore, through the rolling window analysis, the Hurst exponent dynamic trends indicate that COVID-19 has a clear influence on the cross-correlation between movie market and financial market.

Dynamic spillovers of geopolitical risks and gold prices: New evidence from 18 emerging economies

Based on the spillover index model, we study the dynamic spillover effects between the geopolitical risks of 18 emerging countries and gold prices from February 1985 to June 2019. The results indicate that there is a significant spillover effect between geopolitical risks and gold prices, and the net spillover comes from the geopolitical risks on gold prices. Among 18 emerging countries, China, Ukraine, Saudi Arabia, Venezuela, and Argentina are net contributors, with China being the largest net contributor. A rolling window analysis shows that the spillover effects between geopolitical risks and the gold market present an increasing trend, and it would be strengthened significantly in the short term during major geopolitical events. In addition, we decompose gold prices into three types: gold supply shocks, aggregate demand shocks and gold specific demand shocks, and the phenomenon that spillover effects show considerably different patterns and magnitudes is observed.

Time-frequency connectedness between Asian electricity sectors

This study examines the connectedness of the electricity sector in Asia by employing the connectedness models of Diebold and Yilmaz (2012) and Barunik and Krehlik (2018). Our sample includes the following Asian countries: China, Hong Kong, India, Japan, Malaysia, Pakistan, Philippines, South Korea, Thailand, and Vietnam. Our full sample analysis reveals a strongly connected electricity sector in the sample Asian countries, signifying a substantial influence of Asia’s electric utilities on each other. Moreover, our frequency-based analysis exhibits that the connectedness between electricity sectors in Asia is more pronounced in the short-run compared to the long-run horizon. Further, we explore the time-varying connectedness by employing a rolling window analysis that shows the dynamic nature of connectedness and reveals the significant effect of important events like the Chinese financial crisis and the ongoing pandemic crisis. We also explore the connectedness between Asian electricity utility sectors during turbulent times by employing subsample analyses covering the Chinese financial crisis and Pandemic crisis periods. The subsample analyses show that market-wide uncertainty drives up the connectedness, and it is relatively more pronounced for short-run frequencies and during the pandemic crisis period. The network connectedness analysis suggests that regulators could identify countries that most threaten system stability in the Asian electricity sector.

Market Adaptability and Evolving Predictability of Stock Returns: An Evidence from India

This study attempts to examine the adaptive market hypothesis and evolving predictability of stock returns using four decades of daily data from the Bombay Stock Exchange (BSE) and the National Stock Exchange (NSE) in India. The recent developed automatic portmanteau ratio (AVR) and wild bootstrap automatic variance ratio (WAVR) test are used for analysis. We also estimate both the AVR and WAVR statistics in the rolling window framework to examine evolving predictability. The results revealed that BSE and NSE are informationally inefficient in the weak-form. The results of rolling window analysis suggested that the degree of predictable patterns evolves over the period due to global and regional economic and non-economic events. Further, the study compare which stock market is more efficient and found that NSE is more efficient than BSE. The findings of this study provide essential inputs to investors on trading strategies in dynamic economic situations and policymakers to formulate an appropriate policy that can make the Indian stock markets efficient.

Time‐varying causal nexuses between economic growth and CO2 emissions in G‐7 countries: A bootstrap rolling window approach over 1820–2015

AbstractThis paper reinvestigates the causal nexuses between economic growth and CO2 emissions in G‐7 countries during 1820–2015 for all countries except Japan, for which the data spans from 1950 to 2015. The unit root test developed by Perron, Econometrica: Journal of the Econometric Society, 1989, 57, 1361–1401 and the cointegration test of Johansen et al., The Econometrics Journal, 2000, 3, 216–249 which accommodate structural breaks in trends are used to establish a long‐run equilibrium relationship between the variables. Our empirical analysis suggests a long‐run association between the variables. The bootstrap Granger causality test on the full sample and the sub‐samples with a fixed‐window size is used to resolve the issue of structural changes. The findings indicate that the causal relationships are country‐specific. Consistent with the short‐run stability tests, the rolling window analysis shows that causality between GDP and CO2 emissions is time‐varying. These findings offer important policy recommendations for sustainable growth and low‐carbon economy.

Regional housing price dependency in the UK: A dynamic network approach

The cross-regional dependency in the UK housing market is analysed using regional house price indices. In this article, a network approach based on partial correlations is proposed, along with rolling-window analysis to consider potential time-varying dependency. The results show that house prices in the outer South East region have the strongest influence on regional housing market interactions in the UK. This influence is stronger when the markets are highly interconnected, whereas the house prices in London have the strongest influence when the UK regional housing markets are relatively less connected.

Forecasting seasonal influenza-like illness in South Korea after 2 and 30 weeks using Google Trends and influenza data from Argentina.

We aimed to identify variables for forecasting seasonal and short-term targets for influenza-like illness (ILI) in South Korea, and other input variables through weekly time-series of the variables. We also aimed to suggest prediction models for ILI activity using a seasonal autoregressive integrated moving average, including exogenous variables (SARIMAX) models. We collected ILI, FluNet surveillance data, Google Trends (GT), weather, and air-pollution data from 2010 to 2019, applying cross-correlation analysis to identify the time lag between the two respective time-series. The relationship between ILI in South Korea and the input variables were evaluated with Linear regression models. To validate selected input variables, the autoregressive moving average, including exogenous variables (ARMAX) models were used to forecast seasonal ILI after 2 and 30 weeks with a three-year window for the training set used in the fixed rolling window analysis. Moreover, a final SARIMAX model was constructed. Influenza A virus activity peaks in South Korea were roughly divided between the 51st and the 7th week, while those of influenza B were divided between the 3rd and 14th week. GT showed the highest correlation coefficient with forecasts from a week ahead, and seasonal influenza outbreak patterns in Argentina showed a high correlation with those 30 weeks ahead in South Korea. The prediction models after 2 and 30 weeks using ARMAX models had R2 values of 0.789 and 0.621, respectively, indicating that reference models using only the previous seasonal ILI could be improved. The currently eligible input variables selected by the cross-correlation analysis helped propose short-term and long-term predictions for ILI in Korea. Our findings indicate that influenza surveillance in Argentina can help predict seasonal ILI patterns after 30 weeks in South Korea, and these can help the Korea Centers for Disease Control and Prevention determine vaccine strategies for the next ILI season.

Macroeconomic factors and frequency domain causality between Gold and Silver returns in India

This paper examines the relationship between gold and silver returns in India, using monthly data for the period May 1991 to June 2018. To this end, we employ the recently developed frequency domain rolling-window analysis (which is able to show that transitory high frequency shocks are not equal to permanent low frequency shocks over time), as well as the conditional, partial conditional, difference conditional approaches, in addition to the Toda Yamamoto and frequency domain Granger Causalities methods. Further, the relationship is examined in conditional and unconditional frameworks. To condition the relationship, three macroeconomic variables, namely interest rate, BSE stock index and inflation rate are used as the control variables. The results uncover some interesting predictability patterns that vary along the spectrum. Specifically, by applying the rolling-window analysis, we find mixed results of the causality between the gold and silver markets based on the frequencies of different lengths. Our results provide policy inputs, assist investors and hedgers who wish to invest in these markets by constructing strategies and diversify their portfolios based on different frequencies.

Rare disaster and renewable energy in the USA: new insights from wavelet coherence and rolling-window analysis

The increasing trend of economic and political crises in different parts of the world has made global economies highly vulnerable because of having globally as well as regionally integrated economic systems. In such an environment, switching to alternative energy products, such as renewable energy production, may be devastating. Therefore, the aim of this paper is to provide novel insights for the relationship between rare disaster risks and renewable energy production (REN) of the USA by utilizing the time series monthly data from 1973 to 2016. Using time-varying continuous wavelet power spectrum, the wavelet coherence, and the modified bootstrap rolling-window analysis, the results reveal significant linkages between all the categories of rare disaster risks and renewable energy production. Rare disaster risks and REN are linked with each other, and both the variables have time-varying cyclic and anti-cyclic effects on each other with robust and significant predictability from rare disasters to REN. These findings have novel implications for many stakeholders. For instance, producers of energy may safely switch to renewable energy production since disasters are found to have potential to leave cyclic effect on renewable energy at most.

Fancy Bitcoin and conventional financial assets: Measuring market integration based on connectedness networks

We investigate the relationship between Bitcoin and conventional financial assets from a perspective on the connectedness of asset networks. We adopt the method of measuring connectedness proposed by Diebold and Yilmaz (2009, 2012, and 2014) in a VAR system to study the dynamic interdependence between returns in Bitcoin, stocks, oil, and gold. We find that the connectedness between bitcoin and conventional assets is weak. The separation of positive and negative returns in the Bitcoin market shows the existence of an asymmetric pattern of the spillover effects between Bitcoin and conventional assets. A rolling window analysis finds that although Bitcoin prices experience a rising link to other financial assets, the magnitude is proven to be moderate. However, connectedness via negative returns is much stronger than via positive ones and exhibits a clearly increasing trend in recent periods. Our results in application are generally robust to other popular cryptocurrencies, such as ETH and Ripple. The findings presented in this paper have important implications for financial market participants, policymakers, and researchers in light of projected increases in the adoption of Bitcoin, as well as the rapid development of cryptocurrency.

Refined Measures of Dynamic Connectedness based on Time-Varying Parameter Vector Autoregressions

In this study, we enhance the dynamic connectedness measures originally introduced by Diebold and Yılmaz (2012, 2014) with a time-varying parameter vector autoregressive model (TVP-VAR) which predicates upon a time-varying variance-covariance structure. This framework allows to capture possible changes in the underlying structure of the data in a more flexible and robust manner. Specifically, there is neither a need to arbitrarily set the rolling-window size nor a loss of observations in the calculation of the dynamic measures of connectedness, as no rolling-window analysis is involved. Given that the proposed framework rests on multivariate Kalman filters, it is less sensitive to outliers. Furthermore, we emphasise the merits of this approach by conducting Monte Carlo simulations. We put our framework into practice by investigating dynamic connectedness measures of the four most traded foreign exchange rates, comparing the TVP-VAR results to those obtained from three different rolling-window settings. Finally, we propose uncertainty measures for both TVP-VAR-based and rolling-window VAR-based dynamic connectedness measures.

Do oil prices and exchange rates account for agricultural commodity market spillovers? Evidence from the Diebold and Yilmaz Index

ABSTRACT This paper examines the nature of interconnectedness between the returns of the price of oil and foreign exchange on selected agricultural commodity prices. To do this, the authors leverage the novel methodology of a spillover index developed by Diebold and Yilmaz (2012) that reports predictive directional measurement of volatility spillovers. International Journal of Forecasting 28, no. 1: 57–66) that reports: (i) Net spillovers; (ii) Directional spillovers; (iii) Pairwise net spillovers; and (iv) Total spillover indices. This study also captures all secular and cyclical movements with the aid of rolling window analysis to ensure the robustness of the estimations. Empirical analyses are constructed based on monthly realised frequency data from 2006M1 to 2016M7. The empirical analysis from the full sample size shows that rice, sorghum, price inflation, a nominal effective exchange rate and oil price display weak pass-through among the investigated variables while banana, cocoa, groundnut, maize, soybean and wheat are net transmitters of spillover. Based on these revelations, several policy prescriptions for the agricultural commodity markets and their diverse responses to either exchange rate fluctuations or a dwindling oil price are suggested for Nigeria.

Spillover dynamics across price inflation and selected agricultural commodity prices

This article contributes to the existing empirical literature by examining the spillovers across price inflation and agricultural commodity prices for the case of Nigeria. To achieve this objective, we employ the Diebold and Yilmaz (Int J Forecast 28(1):57–66, 2012) spillover index. Subsequently, we examine the directional spillover, total spillover, and net spillover indexes. Further analysis to capture cyclical and secular movements was addressed with 40 months of subsamples via the rolling window analysis. Our empirical results, based on the monthly frequency data from January 2006 to July 2016 show that the total spillover effect was about 75%. This suggests a high interconnectedness of the selected agricultural commodity prices and inflation. Further empirical findings shows that inflation, sorghum, soybeans, and wheat were net receivers while cocoa, barley, groundnut, maize, rice were net givers. We find a negative net spillover for price inflation, implying a net positive spillover from commodity prices to price inflation. Based on these outcomes, several inherent policy implications for the government administrators, farmers, investors and all stakeholders abound. For instance, the need for government officials to insulate the agricultural market from externalities for optimum prices stability is pertinent.

Predictability of precious metals and adaptive market hypothesis

PurposeThe purpose of this paper is to boost the existing literature on adaptive market hypothesis (AMH) as it first time links predictability of gold, silver and metal returns with AMH which permits the predictability of returns to vary over time.Design/methodology/approachTo know whether commodity (gold, silver and metal) market is efficient or not, the commodity returns are observed by using appropriate linear time series tests (variance ratio test, runs test and auto-correlation test). To capture the varying efficiency of three commodities, the study employs subsamples of five years and all sub-samples are exposed to linear econometric tests to reveal how market efficiency (independency of returns) has behaved over time.FindingsIt is found that the commodity market (gold, silver and metal) is adaptive because fluctuation is observed in the market efficiency. Returns of all three commodities go under the periods of efficiency and inefficiency. Thus, AMH is the better description of behavior of commodity markets than traditional efficient market hypothesis.Research limitations/implicationsChoice of sub-sample in the study is the first limitation as the authors employ a sub-sample comprising five years. Second, commission, fee and taxes (transection cost) are ignored in the study. Finally, the results are reported on the basis of linear econometric tests. In future, longer time period sub-sample analysis is suggested by the study to explore the varying nature of the commodities. Moreover, rolling window analysis may be a more appropriate method to elucidate the idea of AMH in further research. It is further suggested that the method used in the study could be helpful and adapted to examine other commodities (metal and agriculture), bonds and equity markets around the world.Practical implicationsThe study will provide a better investment model which can enable the investors to seek more returns in future. Moreover, this research can be extended to explore multiple issues like adaptive behavior of returns from crypto currencies, bonds, stocks and real estate investment trusts.Social implicationsAs all the linear tests reveal that almost all the commodities show inefficient behavior in full sample period, it is clear that past prices widely would be helpful to predict the future prices at NYSE; furthermore, investors can use the time-varying information to reduce the risk of investment at NYSE. The study is helpful for individual investors as well as portfolio managers and brokers to forecast the prices on the bases of findings.Originality/valueThe paper identifies the need to study why behavior of commodity returns varies over time.

Stock market trading volumes and economic uncertainty dependence: before and during Sino-U.S. trade friction

This article mainly studies the interaction between the economic uncertainty and stock market trading volumes changes before and during Sino-U.S. trade friction using multifractal detrended fluctuation analysis (M.F.-D.F.A.) and multifractal detrended cross-correlation analysis (M.F.-D.C.C.A.). Our research aims to reveal whether the economic uncertainty increased by Sino-U.S. trade friction affects stock market trading volume more susceptible, as well as how policymaker strengthen risk management and maintain financial stability. The results show that the dynamic volatility linkages between economic uncertainty and stock market trading volumes changes are multifractal, and the cross-correlation of volatility linkages are anti-persistent. Through the rolling-windows analysis, we also find that the economic uncertainty and trading volumes are anti-persistent dynamic cross-correlated. This means that while economic uncertainty increases, trading volume decreases. Besides, Sino-U.S. trade friction has impact on the cross-correlated behaviour significantly, suggesting that stock markets’ risks are relatively large and trading volumes changes are more susceptible by economic uncertainty during Sino-U.S. trade friction in the U.S. Our study complements existing literature about the stock markets trading volumes and economic uncertainty dependence relationship by multifractal theory’s methods. The overall findings imply that the increased economic uncertainty caused by Sino-U.S. trade friction exacerbates financial risks, which are useful for policymakers and investors.

Does herding behavior exist in cryptocurrency market?

This paper investigates the existence of herding behavior in cryptocurrencies market. Using data of the 20 large cryptocurrencies and MV Index Solution Crypto Compare Digital Assets for large cap index, we found no evidence of herding behavior using cross-sectional absolute standard deviation estimation. However, by applying a rolling window analysis, the results show significant herding behavior, which varies over time. Finally, we find an inverse relationship between herding behavior and the Bloomberg consumer comfort index which means that when traders are less comfortable they prefer to ignore their expectations and follow the market performance.