Time-varying Autoregressive Research Articles

Co-movement of Oil and Stock Markets During COVID-19: Evidence from the Gulf Corporation Council

This study aims to document the impact of oil price variations on generating stock returns in Gulf Cooperation Council (GCC) markets during the COVID-19 era. It documents the pandemic era results from January 2020 to October 2022 by employing cointegration, Granger causality, and time-varying coefficient-vector autoregression techniques on daily data. The findings suggest two-way causality between all stock indices and the oil market. Time-dependent relationships were observed during the review period. The findings (based on regression and variance decomposition) indicate that although the reliance of the GCC stock markets was not on oil alone, oil had a significant impact during the study period. It is recommended that investors not consider diversifying portfolios in GCC stocks and oil markets to optimize benefits. The findings are expected to enhance the understanding of academics, market players, regulators, and investors regarding relationships among GCC stocks and oil markets. This study contributes to the literature by documenting the impact of the oil market on stocks during an abnormal period of the COVID-19 pandemic, considering time-varying parameters in a net oil exporting region.

Sparse-structured time-varying parameter vector autoregression for high-dimensional network connectedness measurement

This paper proposes a new sparse-structured time-varying parameter vector autoregression (SS-TVP-VAR) model to effectively measure high-dimensional network connectedness of financial enterprises. Previous VAR based network connectedness methods require high computational cost and are not feasible for high-dimensional networks of a large number of enterprises. The proposed SS-TVP-VAR not only reduces computational cost but also selects key coefficients when building the VAR model. As a result, it produces a more adaptive and effective connectedness to measure systemic risk caused by extreme events. Studies on 57 listed major financial enterprises in China mainland during a 15-year period show that the proposed SS-TVP-VAR achieves reliable and useful results, and it successfully identifies all the major market events during this time.

Islamic cryptocurrency integration for enhanced sustainable finance: evidence from time-frequency volatility transmission investigation

ABSTRACT This study explores the potential of Islamic gold-backed cryptocurrencies in sustainable finance by focusing on volatility transmission between these cryptocurrencies and conventional digital assets like Bitcoin and PaxGold. Using a Time-Varying Parameter Vector Autoregression (TVP-VAR) approach, we analyze data from December 2019 to July 2023, including the COVID-19 pandemic period. Findings reveal a complex volatility network, with Bitcoin and HelloGold as major transmitters of spillover shocks, while X8X and PaxGold mainly serve as net recipients. Notable interactions between Bitcoin and Islamic gold-backed cryptocurrency markets show short-lived pairwise volatility interactions. Islamic gold-backed cryptocurrencies, net recipients of volatility, absorb market shocks, making them suitable for hedging against volatility. This can aid in diversifying portfolios, mitigating adverse market impacts, and informing risk management and regulatory approaches. Our research emphasizes the importance of Islamic finance in ethical and sustainable investment within the evolving cryptocurrency market.

A novel method to select time-varying multivariate time series models for the surveillance of infectious diseases

BackgroundDescribing the transmission dynamics of infectious diseases across different regions is crucial for effective disease surveillance. The multivariate time series (MTS) model has been widely adopted for constructing cross-regional infectious disease transmission networks due to its strengths in interpretability and predictive performance. Nevertheless, the assumption of constant parameters frequently disregards the dynamic shifts in disease transmission rates, thereby compromising the accuracy of early warnings. This study investigated the applicability of time-varying MTS models in multi-regional infectious disease monitoring and explored strategies for model selection.MethodsThis study focused on two prominent time-varying MTS models: the time-varying parameter-stochastic volatility-vector autoregression (TVP-SV-VAR) model and the time-varying VAR model using the generalized additive framework (tvvarGAM), and intended to explore and verify their applicable conditions for the surveillance of infectious diseases. For the first time, this study proposed the time delay coefficient and spatial sparsity indicators for model selection. These indicators quantify the temporal lags and spatial distribution of infectious disease data, respectively. Simulation study adopted from real-world infectious disease surveillance was carried out to compare model performances under various scenarios of spatio-temporal variation as well as random volatility. Meanwhile, we illustrated how the modelling process could help the surveillance of infectious diseases with an application to the influenza-like case in Sichuan Province, China.ResultsWhen the spatio-temporal variation was small (time delay coefficient: 0.1–0.2, spatial sparsity:0.1–0.3), the TVP-SV-VAR model was superior with smaller fitting residuals and standard errors of parameter estimation than those of the tvvarGAM model. In contrast, the tvvarGAM model was preferable when the spatio-temporal variation increased (time delay coefficient: 0.2–0.3, spatial sparsity: 0.6–0.9).ConclusionThis study emphasized the importance of considering spatio-temporal variations when selecting appropriate models for infectious disease surveillance. By incorporating our novel indicators—the time delay coefficient and spatial sparsity—into the model selection process, the study could enhance the accuracy and effectiveness of infectious disease monitoring efforts. This approach was not only valuable in the context of this study, but also has broader implications for improving time-varying MTS analyses in various applications.

Contagion effect between fuel fossil energies and agricultural commodity markets and portfolio management implications

This study investigates the dynamic spillover effects between fossil fuels (Brent oil and natural gas) and major agricultural markets (maize, soybeans, and wheat) using a time-varying parameter vector autoregression (TVP-VAR) model. Additionally, we quantify the cumulative return of the investment portfolio, optimal weights, time-varying risk coverage, and hedging effectiveness of bivariate commodity portfolios using the Minimum Variance Portfolio (MVP), Minimum Correlation Portfolio (MCP), Minimum Connectedness Portfolio (MCoP) under bearish, tranquil, and bullish market conditions. The results show that maize and natural gas are the main net receivers of spillovers, whereas wheat and Brent oil are the main net transmitters of shocks to the network, respectively. Our findings of optimal weights in multivariate portfolio suggest more weights for agri-commodities against fossil fuels in all three modes of market. In addition, the optimal weights of bivariate portfolio analysis show that the most optimal portfolio which has the highest HE value is related to natural gas pairs with agri-commodities. Moreover, depending on the status of the market; in an average and bearish modes, the highest optimal weight is related to maize/natural gas. However, in the bullish market, the highest optimal weights are related to natural gas/Brent oil pair. Furthermore, the results of optimal bivariate hedge ratios show that natural gas/wheat (wheat/maize and vice versa) portfolio is the cheapest (the most expensive) investment strategy in all modes of market and based on MVP and MCoP approaches.

China’s GDP-at-Risk: Real-Time Monitoring, Risk Tracing, and Macroeconomic Policy Effects

Timely monitoring GDP-at-risk and tracing economic downside risk sources can help establish effective risk warning and prevention systems. This study constructs a probability distribution for China’s economic growth with skewness determined by a multidimensional predictor information set of macro fundamentals. Such a treatment allows us to identify changing drivers of economic downside risks during monitoring GDP-at-risk’s dynamic evolutionary path. We also employ a time-varying parameter vector autoregression model with random volatility to explore the heterogeneous impacts of different macroeconomic policy instruments on economic slowdowns. Our results provide empirical support for macroeconomic management and policy formulation in emerging markets. We reach three conclusions. First, the dynamics of GDP-at-risk exhibit significant event-driven characteristics, and economic downside risk increases significantly under the influence of extreme events. Moreover, the probability distribution of economic growth is asymmetric--as the downside risk of the economy increases, its upside potential increases disproportionately. Second, the time-varying risk trace of GDP-at-risk shows that the contribution of financial conditions and local government debt to economic downside risk declines. The importance of the risk-driving role of housing price growth gradually increases, suggesting that China’s property prices can provide more valuable early warning information about future growth risk, allowing time for precise preventive measures. Nevertheless, interest rates and inflation as risk divers have consistently minimal impacts. Third, the heterogeneity impulse response function of GDP-at-risk suggests that quantity-based monetary policy and fiscal policy can manage economic downside risks in the short run. In contrast, price-based monetary policy can curb economic overheating and reduce downside risks in the medium to long term. Therefore, the effect of price-based monetary policy is more sustainable in China.

Leadership shift in the global soybean market: Dynamic connectedness approach (TVP-VAR)

The price transmission in international soybean market has been extensively examined. However, recent econometric advancements have enabled the application of dynamic connectedness methodology as outlined by Antonakakis and Gabauer (2017) [1], which is based on a Time-varying Parameter Vector Autoregressive (TVP-VAR) model. This approach captures the time-varying connectedness of time series, considering potential risk shock emitters and receivers. The connectedness index (Diebold and Yilmaz, Jan. 2012) [2] was developed using Generalized Forecast Error Variance Decomposition (GFEVD) and the Generalized Impulse Response Function (GIRF) (Koop et al., Sep. 1996; Pesaran and Shin, Jan. 1998) [3,4]. This study aims to understand the dynamic connectedness and price leadership. The research examined markets including the US Soybean futures market (Chicago Board of Trade), Rotterdam Port spot market representing the European soybean market, Paranaguá port representing Brazil, Argentina represented by Rosario Futures and Spot, and the Chinese domestic spot market and the Dalian futures on behalf of China. The research spanned approximately ten years, from September 2009 to May 2019. The findings suggest that the soybean market has reached a high level of maturity, able to withstand exogenous shocks for at least the past seven years. The net pairwise directional connectedness revealed dynamic and bidirectional causality. The dynamic connectedness index showed a highly connected and developed market in the West (Chicago Futures, Rotterdam, Paranaguá, and Rosario Futures and Spot). However, the connectedness between Western and Eastern markets was relatively low, indicating some level of market isolation. Furthermore, the pairwise connectedness index between Eastern market (China spot and Dalian Futures) was also considerable low. Lastly, Paranaguá and Rosario overtook Rotterdam as price-leading markets and were identified as the primary net transmitters of shocks, indirectly implying causality. Chicago, Rosario, and Paranaguá formed a triumvirate leading international prices, while Rotterdam adopted a secondary leading position, deviating from its historical role as the price leader.

Dynamic interconnectedness of economic confidence, energy prices, and interest rates: Insights from the euro area

This work examines the time-varying interlinkages among economic confidence, energy prices, geopolitical stress, and short/long-term interest rates in the Euro Area. Our research meticulously explores the interplay between economic confidence and various determinants, including financial indicators, geopolitical stress incidents, and energy prices. Employing innovative approaches such as the time-varying parameter vector autoregression (TVP-VAR) time and frequency-domain connectedness, we uncover the nuanced relationships between economic confidence, financial indicators, and energy prices. We illuminate the systemic nature of shock transmission in the Euro Area, identifying key net transmitters and recipients of shocks, with short-term interconnectedness emerging as a dominant feature, especially during pivotal events such as the global financial crisis, the COVID-19 pandemic, and geopolitical conflicts. Our empirical findings can be summarized as follows: First, both the time and frequency-domain connectedness indexes correctly associate with major financial/geopolitical events. Second, BCI and CCI respond to the GFC asymmetrically. Third, Brent and short/long-term interest rates are the net transmitters of shocks on average. Fourth, there is a considerable augmentation in return spillovers during the period characterized by the pandemic crisis compared to the GFC. Finally, our findings for frequency-dependent connectedness networks indicate that the market is particularly susceptible to short-term shocks. This study has significant ramifications for investors, market players, and policymakers.

Tracking trend output using expectations data

Abstract This article proposes a new approach to measuring trend output that exploits survey data on expectations to distinguish the effects of permanent and transitory shocks and to track the time-variation in the processes underlying the determination of output. The approach is illustrated using measures of output expectations and output uncertainties based on a business survey conducted for UK manufacturing. The measures are employed in a time-varying vector autoregression (VAR) to track trend output and to provide a compelling characterization of the output fluctuations in UK manufacturing over the last 20 years.

Exploring the dynamic connections between oil price shocks and bond yields in developed nations: A TVP-SVAR-SV approach

This paper investigates the impact of oil price changes on bond yields across nine developed nations, focusing on how demand, supply, and oil-specific demand shocks influence bond spreads. Using a Time-Varying Parameter Structural Vector Autoregression with Stochastic Volatility (TVP-SVAR-SV) model, we find that bond yields are short-term responsive to oil price fluctuations, indicating efficient domestic bond markets. However, we observe distinct country-specific reactions, particularly in the US and the UK. In the US, supply shocks cause temporary declines in bond yields, while demand shocks reduce bond yields in both countries. Oil-specific demand shocks have cyclical impacts. Unlike the global financial crisis, the COVID-19 pandemic uniquely influenced bond yields, with minimal short-term effects in Canada, France, Spain, and the US post-COVID-19, contrasting with the severe impacts during the financial crisis. These findings suggest investors should consider the varying impact of oil shocks on bond yields across countries, and policymakers should monitor and address the potential implications of oil price fluctuations for financial stability and market efficiency.

Overlooked carbon sinks in the house building industry and the influential factors

The study aims to establish a carbon sink measurement model for the house building industry and analyze the time-varying characteristics of influential factors. China's house building industry exhibited an increasing trend in carbon sink, from 0.336 billion tons in 2000 to 234.1 million tons in 2020. This study defines the carbon sink coefficient of the house building industry to understand its carbon sink capacity. From 2000 to 2020, the average carbon dioxide absorption per square meter of floor space annually, throughout its life cycle, is 2.09 kg. Factors influencing the carbon sink were analyzed using Logarithmic Mean Divisia Index (LMDI) model. The most influential factor was resource consumption intensity, followed by value dependence level and output scale effect. Employing Time-Varying Parameter Vector Autoregression (TVP-VAR), this study examines the impact of macro factors on carbon sinks at various lead times and specific points. Resource consumption intensity exhibits significant time-varying effects. Value dependence level negatively influences, primarily in the medium and long terms, while the output scale effect consistently shows a positive influence, mainly in the long term. Scenario forecasting suggests that by 2035, under low-carbon transition scenarios, carbon sinks could increase by 8.28% and 17.26% compared to baseline and high-carbon crude scenarios, respectively. To foster this growth, prolonging the service life of houses, enhancing construction site management and waste treatment, and employing higher value-added production factors are recommended.

The impact of COVID-19 on global financial markets: A multiscale volatility spillover analysis

This paper presents a comprehensive multiscale analysis of volatility spillover mechanisms among diverse financial markets utilizing a novel combination of Normalized Adaptive Multivariate Empirical Mode Decomposition (NA-MEMD) and Time-Varying Parameter Vector Autoregressive (TVP-VAR) model. Our analysis spans Bitcoin, crude oil, gold, foreign exchange, emerging stock markets, and developed stock markets, examining their respective roles in global volatility transmission. Utilizing the NA-MEMD methodology and a time-varying volatility spillover index, we find that in the short-term, Bitcoin and crude oil primarily act as volatility receivers, while other markets serve as volatility transmitters. Over the long-term, gold and Bitcoin consistently act as volatility receivers, with Bitcoin and gold showing evident safe-haven effects, while the remaining markets primarily function as volatility transmitters. Particularly post-COVID-19, developed stock markets emerge as significant transmitters of volatility, largely directed towards the crude oil market. By evaluating and comparing risk transmission patterns among financial markets before and during the COVID-19 pandemic, our findings offer valuable insights that can assist policymakers and other stakeholders in managing financial uncertainty in times of global crises.

A new method for measuring financial resilience

This paper proposes a new model to measure financial resilience from the perspective of external risk shocks, which composes of three submodels, namely, the dynamic factor model, the time-varying parameter vector autoregression (TVP-VAR) model, and the resilience characteristic measurement model with two dimensions: absorption intensity and absorption duration. At the theoretical level, we simulate and analyze the changing paths of financial resilience under different scenarios. At the empirical level, we apply the model to study the resilience of the UK financial market. The results indicate that the UK financial resilience fluctuations exhibit phased characteristics and there is a clear inverse relationship between absorption intensity and absorption duration. Notably, periods of low resilience often coincide with specific risk events.

The trend of digital finance: unveiling the multidimensional network of cryptocurrency risk propagation

ABSTRACT This study primarily explores the mechanisms of risk propagation among cryptocurrencies, unveiling for the first time the frequency dimension of risk propagation within the cryptocurrency market and identifying the role of oscillation frequency in this process. By employing Variational Mode Decomposition (VMD) and the DY spillover matrix to construct a complex network, the paper analyzes the frequency dimension risk propagation mechanisms of nine major cryptocurrencies from 2017 to 2023. Key findings include the significant risk propagation capabilities of Ethereum (ETH) and Bitcoin (BTC) during periods of high market volatility, while stablecoins such as Tether and USD Coin exhibit minimal risk propagation ability. Additionally, the characteristics of cryptocurrency risk propagation have been enhanced following the COVID-19 pandemic. Overall, the risk propagation of most cryptocurrencies is primarily realized through high-frequency oscillations. The robustness of the conclusions is verified using the Time-Varying Parameter Vector Autoregressive (TVP-VAR) model. The results are significant for understanding the dynamic characteristics of the cryptocurrency market, predicting future market risks, and formulating risk management strategies. Furthermore, the methodology and findings of this study provide new perspectives and tools for exploring the risk propagation relationships among cryptocurrencies.

Volatility and returns connectedness between cryptocurrency and China’s financial markets: A TVP-VAR extended joint connectedness approach

We employ a time-varying parameter vector autoregression (TVP-VAR) joint connectedness approach to study the dynamic risk spillover effects between cryptocurrencies and China’s financial market, further exploring the impact of cryptocurrencies on China’s financial market. Our results show that there is asymmetric risk transmission between cryptocurrencies and China’s financial market, and the risk spillover effect is very weak. Specifically, the spillover of cryptocurrencies to China’s financial market is significantly stronger than the spillover of China’s financial market to cryptocurrencies. Cryptocurrencies have a stronger spillover effect to China’s exchange rate and gold. The net spillover effect of cryptocurrencies is weakening over time. Overall, the return spillover impact of cryptocurrencies on China’s financial market is greater than the volatility spillover impact, and the degree of impact of different cryptocurrencies is heterogeneous. The findings of this study have several implications for policymakers and investors.

Does the U.S. export inflation? Evidence from the dynamic inflation spillover between the U.S. and EAGLEs

Given the crucial role of inflation as a key economic barometer, our paper investigates the dynamic inflation spillover between the U.S. and the nine emerging and growth-leading economies (EAGLEs) between 1991M1 and 2020M2. Employing the recently developed time-varying parameter vector autoregressions (TVP-VAR)-based connectedness approach, we find evidence of a moderate inflation spillover across the sample countries at normal condition. We further point out that inflation spillover effects with the U.S. are more pronounced for the emerging markets with higher openness, the net oil-importing emerging markets, and the emerging markets following free-float exchange rate regimes. More importantly, the inflation spillover index among the system rises dramatically to over 70% under extremely inflationary conditions, implying that the transmission of spiral inflation is very high. Additionally, the time-varying analysis shows that the role of the U.S. in the inflation shock transmission with emerging countries varies between being a net inflation-exporter and inflation-importer over times. Finally, an investigation of the drivers of the inflation spillovers reveals that the U.S. dollar, emerging markets' economic policy uncertainty, and bilateral trade are key determinants of the inflation shock transmission among the system. Our findings justify central banks’ actions in decreasing U.S. dollar reserves to safeguard their domestic currencies.

The time-varying effects of geopolitical risk on mutual fund risk taking.

Based on a time-varying parameter vector autoregression model with stochastic volatility (TVP-VAR-SV), this paper investigates the dynamic effects of geopolitical risk on mutual fund risk taking in China across three-time horizons and at three selected time points. Overall, the impulse responses are time-varying and we find a negative effect of geopolitical risk on mutual fund risk taking until 2015, with the short-term effect being the most pronounced, suggesting that when professional investors such as mutual fund managers are faced with the stock valuation uncertainty due to a geopolitical shock, they choose to reduce market risk exposures. After 2015, the short-term effect begins to diminish and gradually turns positive, which could be explained by the fact that with the increasing abundance and diversification of investment instruments, fund managers have more effective investment tools and more sophisticated trading strategies to hedge against geopolitical risk, rather than reducing market risk exposure. Further, we explore the heterogeneous effects of eight types of geopolitical risk and three types of mutual fund. The results indicate that the effect of geopolitical actions is stronger than that of geopolitical threats, while the effect of narrow geopolitical risk is stronger than that of broad geopolitical risk. Moreover, we find that the response of the risk taking of growth funds to the geopolitical risk is weaker than that of balanced and income funds.

A comparison of several cyclo-stationary and fractionally differenced time-varying autoregressive models in runoff simulation and prediction

Due to the far-reaching effects of climate change and human activities, hydrological processes exhibit cyclo-stationarity and long-term memory. Neither the traditional autoregressive (AR) nor the time-varying autoregressive (TVAR) model can comprehensively characterize these runoff characteristics, leading to poor model prediction performance. We propose a novel fractionally differenced TVAR model based on the de-seasonality method (FDTVARD) to better capture the above runoff characteristics and achieve improved runoff simulation and prediction accuracy. We first compare the effects of two covariates, time and lagged target variables, on the performance of the FDTVARD model. The FDTVARD model is compared with the AR model based on the de-seasonality method (ARD), fractionally differenced ARD (FDARD) and TVAR model based on the de-seasonality method (TVARD), and the model performance is evaluated using daily runoff data from seven hydrological stations in the Yellow River Basin, China. The results show that (1) the FDTVARD model has the simplest model form and yields the best simulation and forecasting performance, and the optimal covariate is time. (2) Considering only the time-varying nature of the model coefficients improves the ARD model prediction performance more than considering only the long-term memory of the runoff series. Considering both factors improve the simulation and prediction performance of the ARD model more than considering only one factor. The study findings could facilitate improvements in the prediction accuracy of stochastic hydrological models.

Time-varying interrelations between digitalization and human capital in Vietnam

In this study, we use a time-varying parameter vector autoregression in conjunction with the extended joint connectedness approach (TVPVAR-EJC) to examine the interrelationships among five variables, namely digitalization measured by the number of individuals using the Internet (% of the population), mobile cellular subscriptions (per 100 people), human capital index, CO2 emissions, and GDP, from 1996 to 2019 in Vietnam. Net total directional connectedness of digitalization and other variables suggests that it consistently acts as a net shock transmitter, whereas human capital was found to be a net receiver of corresponding shocks from 1997 to 2019. The emission of CO2 was a net critical receiver in 1998 but remained as a net transmitter for the rest of the period. Pairwise connectedness reveals that digitalization appeared as a net shock transmitter in connection with the human capital index and GDP. In the relationship with human capital, digitalization’s role was critical.

Extant linkages between Shanghai crude oil and US energy futures: Insights from spillovers of higher-order moments

This study is epicentral to analyzing the impact of futures volatility on portfolio and risk management, as extant literature indicates the challenges of using economic variables that fall short of forecasting volatility beyond lagged values. Further, higher moments may be better adaptive to signaling distress during market upheavals. This paper sources data from Bloomberg from March 26, 2018–April 28, 2023, to examine the dynamic spillovers of higher moments among Shanghai International Energy Exchange and US energy futures contracts by constructing realized skewness and kurtosis. Using nonlinear techniques of mutual information and time-varying vector autoregression (TVP-VAR), we show that realized skewness and kurtosis offer significant information on spillover transmission between the two futures markets, primarily through the crises of COVID-19 and the Russia and Ukraine war. Further, we identify that the risks embedded in these future contracts have increased significantly. Our results have important implications for policymakers, investors, and risk managers.