Time-varying Correlation Research Articles

Dynamic functional connectivity analysis based on time-varying partial correlation with a copula-DCC-GARCH model

We suggest a time-varying partial correlation as a statistical measure of dynamic functional connectivity (dFC) in the human brain. Traditional statistical models often assume specific distributions on the measured data such as the Gaussian distribution, which prohibits their application to neuroimaging data analysis. First, we use the copula-based dynamic conditional correlation (DCC), which does not rely on a specific distribution assumption, for estimating time-varying correlation between regions-of-interest (ROIs) of the human brain. Then, we suggest a time-varying partial correlation based on the Gaussian copula-DCC-GARCH model as an effective method for measuring dFC in the human brain. A recursive algorithm is explained for computation of the time-varying partial correlation. Numerical simulation results demonstrate effectiveness of the partial correlation-based methods against pairwise correlation-based methods. In addition, a two-step procedure is described for the inference of sparse dFC structure using functional magnetic resonance imaging (fMRI) data. We illustrate the proposed method by analyzing an fMRI data set of human participants watching a Pixar animated movie. Based on twelve a priori selected brain regions in the cortex, we demonstrate that the proposed method is effective for inferring sparse dFC network structures and robust to noise distribution and a preprocessing step of fMRI data.

Bitcoin and liquidity risk diversification

This paper investigates the potential diversification benefits of adding Bitcoin from a liquidity perspective. Using different multivariate GARCH-Type specifications to model the joint dynamics of the selected financial assets, we find that VARMA (1, 1)-cDCC-GARCH is the best-fit model. By examining the dependence structure between Bitcoin and several financial assets, we find evidence of a low time-varying correlation of liquidity innovations over the period 2014–2019. This finding suggests the existence of a potential gain in diversifying the liquidity risk using Bitcoin instead of traditional assets. The results of the portfolio optimization problem when adding the liquidity constraint show that holding Bitcoin can reduce the liquidity risk under the Mean-Variance-Liquidity framework.

Analysis of Performance Deviation of Wind Power Enterprises in China

Considering increased global emphasis on energy security, low-carbon economy and environmental governance, the proportion of renewable energy will increase in national power grid systems. Wind power generation will play an important role in China's future power systems. Environmental uncertainty will affect the time-varying correlation between carbon efficiency and the performance of the wind power enterprises. Panel data from 2011 to 2018 is used to analyze the development status and existing problems of the wind power industry; further, with the data, the dynamic conditional correlation coefficient is calculated through the DCC-GARCH model, and the breakpoint analysis method is used to analyze the impact of policy and the economic environment on the time-varying correlations between carbon efficiency and the performance of wind power enterprises. The results show that during the China-US trade war, the sudden trade policy changes that increased tariffs led to a sharp rise in interdependence. The relevant polices inspiring the wind power industry are positively correlated with the dynamic conditional correlation coefficient, while the financial performance of alternative fossil fuels is negatively correlated with the dynamic conditional correlation coefficient.

Brexit and the dependence structure among the G7 bank equity markets

The UK referendum in June 2016 on leaving the European Union had a negative impact on banking stocks across the major financial markets. This has left with a question dealing with the effect of UK banking institutions on the systemic risk on a global scale. This paper aims at investigating the changes in the dependence structure between the UK bank equity returns and its counterparts in the G7 economies. The methodology used is based on the GJR-GARCH volatility spillover model that accounts for asymmetry and leverage, and copula for the time-varying correlation structure among G7 banks. Taking the data on bank equity return indices for G7 economies, the results indicate the symmetric dependence structure between the UK and Italian banks and the asymmetric dependence between the UK and the rest of G7 banks. This is due to the simultaneous decline in bank shares prices across the Union. Such results are important constituents for cross-country portfolio diversification.

Nonparametric estimation of time varying correlation coefficient

We propose a new time varying correlation coefficient, which is a local correlation measure of a pair of time series. The time varying correlation coefficient is locally estimated using a nonparametric kernel method. Asymptotic normality of the estimated time varying correlation is established, which allows us to construct statistical methods of confidence interval and hypothesis tests. Finite sample validity of the proposed methods are demonstrated by a Monte–Carlo study. The proposed time varying correlation coefficient method is well illustrated by an analysis of five sets of world major stock price index returns.

Asymmetric Effect and Dynamic Relationships Between Stock Prices and Exchange Rates Volatility

The analysis of time-varying correlation between stock prices and exchange rates in the context of international investments has been well researched in the literature in last few years. In this paper, we study the interdependence of exchange rates and stock prices for seven countries (Canada, Japan, Denmark, Hong Kong, Singapore, Mexico and Brazil). To do so, we both use the DCC-FIEGARCH and FIAPARCH-DCC models during the period spanning from January 1, 2000 until January 1, 2016. The empirical results suggest asymmetric responses in stock prices-exchange rates linkages, a high persistence of the conditional correlation. They also show bidirectional spillovers effects between different series. Moreover, the results indicate different behavior of exchange rates-stock prices nexus during the crisis periods, suggesting the need for specific policy measures. Finally, our findings offer investors, portfolios managers and policymakers insights on international portfolios and monetary.

Going Behind the High Rates of NEETs in Italy and Spain: The Role of Early School Leavers

This paper aims to analyse the determinants of the dynamics of those not in employment, education, or training (NEET) during 2007–2017, focusing primarily on the role played by early school leavers (ESLs). The analysis covers Italy and Spain, because they show the highest levels of NEETs and ESLs in the EU, respectively. The dynamics and causes of NEET status in these two countries are revealed to be similar only in appearance. After the analysis of the theoretical framework for the relationship between NEETs and ESLs, we investigate the NEET composition in terms of inactive and unemployed individuals, and empirically study the relationship between ESLs and NEETs both in terms of the interdependence and dependence of NEETs and ESLs. Furthermore, we study the contemporaneous relationship between NEET and ESL rates in Italy and Spain in terms of correlation. Finally, to understand how ESL status can lead to NEET status, we investigate the dynamic relationship between NEET and ESL rates based on simple regression models of NEET rates on lagged rates of ESLs, as well as on a time-varying correlation model. Results reveal, among other things, that Italy shows a greater ESL influence on NEET rates, especially for men. The main conclusion concerns the need to invest more in education and to adopt measures able to keep more young people in education for a prolonged period. Therefore, particularly in Italy, policymakers should pay more attention to the education system, including aspects linked to the labour market.

Co-movement across european stock and real estate markets

This study questions the reliability of securitized real estate investment trusts (REITs) for reducing risk in European stock investment portfolio during the global and European crisis periods. It considers co-movement and portfolio management by analyzing the time-varying correlation, hedging ratios, and portfolio weights. Using daily data from January 6, 2003 to April 11, 2018, the results show significant shift in correlation coefficients between the two assets under financial and economic stress. Consequently, potential diversification benefits appear limited when investing across REITs and stock markets. During crisis periods, investors are subject to a high cost for rebalancing their positions in REITs to mitigate stock portfolio risk. Further analyses confirm the inability of REITs to play hedge and safe-haven roles against stock market downturns. Our findings provide useful insights for investors and portfolio managers to formulate trading strategies, determine asset allocation, and assemble optimal portfolios.

Exogenous shocks, dynamic correlations, and portfolio risk management for the Asian emerging and other global developed and emerging stock markets

ABSTRACT This study examines the potential influence of exogenous shocks on time-varying correlations and portfolio strategies between the Asian emerging and other global stock markets including developed and other emerging markets. Using the ARMA-cDCC-FIEGARCH model with and without exogenous shocks, our results highlight the usefulness of including other global stock assets in the traditional portfolio for Asian emerging market investors. However, investors have limited opportunities to diversify their assets during the global financial crisis. Moreover, the shocks from the U.S. stock market have a greater influence on global stock markets compared to that from U.S. economic policy. Fortunately, the model with exogenous shocks improves its accuracy, which plays the same role of controlling structural breaks in the model. More importantly, incorporating exogenous shocks in our model also provides better value-at-risk performance results and hedging effectiveness. These results have several important implications for investors, researchers, and policymakers.

Implied volatility relationships between crude oil and the U.S. stock markets: Dynamic correlation and spillover effects

This paper investigates the dynamic correlation and risk transmission between the oil market and the U.S. stock market, using the respective implied volatility indices published by the Chicago Board Options Exchange. The results indicate that, first, there is a significant positive time-varying correlation between oil and stock implied volatility returns. Second, during the global financial crisis, the correlation between oil and stock markets increases significantly. Third, there is a significant bidirectional implied volatility spillover between the oil and stock markets. Insights gleaned from the findings in this study could project energy and monetary policy implications. Monetary and/or energy policy changes could impact the predicted linkage mechanism between these two markets, which can be further leveraged to forecast the market's future volatility.

In Search of Gold: Insights From Google Searches on Portfolio Hedging Effectiveness

Online investor attention, measured by Google searches, may provide valuable information for the assessment of the co-movement between financial assets. In our empirical analysis, we draw upon an extension of the dynamic conditional correlation model and find that online investor attention is a statistically significant determinant of the time-varying correlations between stocks and gold-, silver-, and crude oil future contracts. Our results, relying on daily data over eleven years, suggest that market participants may enrich asset allocation strategies by means of online investor attention. Taking a practical point of view, we demonstrate that the informational content of Google searches increases the hedging effectiveness of combined portfolios.

Contagion in Futures FOREX Markets for the Post- Global Financial Crisis: A Multivariate FIGARCHcDCC Approach

This paper seeks to investigate the time-varying conditional correlations to the futures FOREX market returns. We employ a dynamic conditional correlation (DCC) Generalized ARCH (GARCH) model to find potential contagion effects among the markets. The under investigation period is 2014-2019. We focus on four major futures FOREX markets namely JPY/USD, KRW/USD, EUR/USD and INR/USD. The empirical results show an increase in conditional correlation or contagion for all the pairsof future FOREX markets. Based on the dynamic conditional correlations, KRW/USD seems to be the safest futures FOREX market. The results are of interest to policymakers who provide regulations for the futures FOREX markets.
 JEL Classification Codes: C58, C61, G11, G15

Dynamic functional connectivity and graph theory metrics in a rat model of temporal lobe epilepsy reveal a preference for brain states with a lower functional connectivity, segregation and integration

Epilepsy is a neurological disorder characterized by recurrent epileptic seizures. The involvement of abnormal functional brain networks in the development of epilepsy and its comorbidities has been demonstrated by electrophysiological and neuroimaging studies in patients with epilepsy. This longitudinal study investigated changes in dynamic functional connectivity (dFC) and network topology during the development of epilepsy using the intraperitoneal kainic acid (IPKA) rat model of temporal lobe epilepsy (TLE). Resting state functional magnetic resonance images (rsfMRI) of 20 IPKA animals and 7 healthy control animals were acquired before and 1, 3, 6, 10 and 16 weeks after status epilepticus (SE) under medetomidine anaesthesia using a 7 T MRI system. Starting from 17 weeks post-SE, hippocampal EEG was recorded to determine the mean daily seizure frequency of each animal. Dynamic FC was assessed by calculating the correlation matrices between fMRI time series of predefined regions of interest within a sliding window of 50 s using a step length of 2 s. The matrices were classified into 6 FC states, each characterized by a correlation matrix, using k-means clustering. In addition, several time-variable graph theoretical network metrics were calculated from the time-varying correlation matrices and classified into 6 states of functional network topology, each characterized by a combination of network metrics. Our results showed that FC states with a lower mean functional connectivity, lower segregation and integration occurred more often in IPKA animals compared to control animals. Functional connectivity also became less variable during epileptogenesis. In addition, average daily seizure frequency was positively correlated with percentage dwell time (i.e. how often a state occurs) in states with high mean functional connectivity, high segregation and integration, and with the number of transitions between states, while negatively correlated with percentage dwell time in states with a low mean functional connectivity, low segregation and low integration. This indicates that animals that dwell in states of higher functional connectivity, higher segregation and higher integration, and that switch more often between states, have more seizures.

Flexible copula models with dynamic dependence and application to financial data

A new class of copula models with dynamic dependence is introduced; it can be used when one can assume that there exist a common latent factor that affects all of the observed variables. Conditional on this factor, the distribution of these variables is given by the Gaussian copula with a time-varying correlation matrix, and some observed driving variables can be used to model dynamic correlations. This structure allows one to build flexible and parsimonious models for multivariate data with non-Gaussian dependence that changes over time. The model is computationally tractable in high dimensions and the numerical maximum likelihood estimation is feasible. The proposed class of models is applied to analyze three financial data sets of bond yields, CDS spreads and stock returns. The estimated model is used to construct projected distributions and, for the bond yield and CDS spread datasets, compute the expected maximum number of investments in distress under different scenarios.

Oil price shocks and EMU sovereign yield spreads

This study examines for the first time the relationship among the oil price shocks and the sovereign yield spreads in the EMU (which is collectively the largest oil-importer of the world), in a time-varying environment. In particular, we examine the time-varying correlation between oil price shocks and the 10-year sovereign yield spread of core and periphery countries in the EMU, by employing a scalar-BEKK framework. The main findings reveal that the correlations between sovereign yield spreads and oil price shocks are indeed time-varying and are influenced by specific economic and geopolitical events that took place during the study period. Furthermore, even though the correlation patterns are constantly low or zero prior to the Great Recession, a change is revealed in the post-2008 period, when correlations become moderate and more volatile. Finally, we do not observe noteworthy differences in the correlation behaviour between core and periphery countries to different oil price shocks. The findings of this study are particularly useful and provide valuable information to marketplace participants.

The Time-Varying Correlation between Crude oil Future and USA Bond Markets During 2005-2020: Evidence from a DCC-GARCH Model

In this paper, we examine potential time-varying correlations between crude oil future and USA bond markets. We employ a dynamic conditional correlation (DCC) multivariate GARCH model in order to quantify potential contagion effects between the markets for the period 2005-2020. We divide the period in two sub-period to make the empirical analysis easier. Empirical results reveal increased conditional correlation in the first sub-period (2005-2012) and no contagion in the second sub-period (2012-2020). Results are of interest to investors, who invest long-term into the under investigation financial markets, as well as, to policymakers, who provide regulations for the under investigation derivate market.

Prelivanje nestabilnosti i efekat zaraze između tržišta evro dolarskih fjučersa i bezkuponskih obveznica - dokazi iz Italije

This paper examines the time-varying conditional correlations between the Eurodollar futures market and the zero coupons of Banca Fideuram. We apply a bivariate dynamic conditional correlation (DCC) GARCH model in order to capture potential contagion effects between the markets for the period 2005-2017. Empirical results reveal contagion during the under-investigation period regarding the twenty-one bivariate models, showing that the Eurodollar futures market has a major impact on the zero coupons of Banca Fideuram. Findings have crucial implications for policymakers who provide regulations for the above-mentioned derivative markets.

Explaining the stock-stock, bond-bond and stock-bond correlation across countries

This paper examines the behaviour of the same asset-cross country and cross-asset same country correlations for stocks and bond for four (Germany, Japan, UK, USA) major economies. Using the realised volatility methodology to construct time-varying correlations, the results reveal that rising same asset correlations occur when cross-asset correlations fall. While there is evidence of segmentation of Japanese assets within international markets. We seek to explain the movement in correlations and note that the variables that exhibit a positive predictive relation for the stock-bond correlation, exhibit a negative predictive relation for the stock-stock and bond-bond correlations and that this is linked to economic conditions. Within this, four variables (inflation, stock returns, consumer sentiment and purchasing managers index) exhibit consistent significance across the regressions. Using these variables, we construct a correlation indicator variable that is used to construct a switching portfolio. This constructed portfolio constructed outperforms buy-and-hold alternatives.

An empirical examination of correlation dynamics between commodity and equity derivative indices: evidence from India using DCC-GARCH models

This empirical study investigates the time-varying co-movements and volatility linkages between equity-commodity indices and inter-commodity indices. This paper deploys DCC-GARCH framework. Three versions of GARCH namely standard, threshold and exponential with both symmetric and asymmetric versions of dynamic conditional correlations (DCC) have been used. This study considers equity index Nifty 50, non-agricultural commodity indices MCX Energy and MCX Metal along with agricultural commodity indices Dhaanya and MCX Agri. Our results revealed that the combined portfolio of commodities-equity has low or negative correlations, which provide better diversification property rather than a portfolio without commodities. However, we notice a steep decline in time-varying correlations of equity with non-agricultural commodity portfolio since 2013. The mixed portfolio of the agricultural commodity with non-agricultural commodity can also offer diversifying benefits. We found high volatility shifts in time-varying co-movements of Nifty 50-MCX Agri pair. On the other hand, we observed an increasing trend in the co-movements of Nifty 50-Dhaanya, it indicates the interdependency of these two markets. Non-agricultural commodity pair (MCX Energy-MCX Metal) and agricultural commodity pair (MCX Agri-Dhaanya) fail to offer better-diversifying properties. This study provides an essential insight to policymakers, portfolio managers, domestic and international investors, risk analysts and financial researchers in an emerging market.

Revisiting the Integration of China Into the World Crude Oil Market: The Role of Structural Breaks

The China’s crude oil futures market (INE market), as it was first launched in late March of 2018, quickly draws much attention from global investors. In reference to the high frequency data, this research explores how well this new product reacts efficiently to international influences and to what extent it can be integrated with traditional benchmarks such as WTI and Brent. The multivariate GARCH models are employed to capture the cross-market time-varying correlations, return and volatility spillovers, which are modified by incorporating the detected structural breaks in the return dynamics to improve the accuracy of model estimates. Empirical results indicate a strong integration of INE market with these international benchmarks. A high but time-varying correlation is observed with recurring highs around 0.7. Spillover effects have included significant bidirectional return and volatility spillovers between the INE and the international benchmark markets. Secondly, INE market appears to interact better with the Brent market than with the WTI market. Thirdly, structural breaks can influence correlations, the portfolio weights and hedge ratios. Lastly, the correlation between crude oil futures markets decreases significantly during the periods when structural breaks caused by economic and/or geopolitical events are identified. These findings have important implications in policy makings and economic decisions on portfolio management and hedging strategies.