Could volatile cryptocurrency stimulate systemic risks in the energy sector? Evidence from novel connectedness models

Abstract By identifying the connectedness of seven indicators from January 1, 2019, to June 13, 2022, we choose an extended joint connectedness approach to a vector autoregression model with time-varying parameter (TVP-VAR) to analyze interlinkages between Crypto Volatility (CV) and Energy Volatility (EV). Our findings show that the COVID-19 outbreak seems to have an impact on the dynamic connectedness of the whole system, which peaks at about 60% toward the end of 2019. According to net total directional connectedness over a quantile, throughout the 2020–2022 timeframe, natural gas and crude oil are net shock transmitters, while the CV, clean energy, solar energy, and green bonds consistently receive all other indicators. Specifically, pairwise connectedness indicates that the CV appears to be a net transmitter of shocks to all energy indicators before the COVID-19 outbreak but acts as a net receiver of shocks from clean energy, wind energy, and green bonds in late 2020. The CV mostly has spillover effects on green bonds. The primary net transmitter of shocks to the Crypto market is crude oil. Our findings are critical in helping investors and authorities design the most effective policies to lessen the vulnerabilities of these indicators and reduce the spread of risk or uncertainty.


Introduction
As measured by returns and volatility spillovers, the intermarket linkage factor plays a crucial role in international finance, with significant implications for portfolio allocation and hedging decisions [1].Technological advancements, financial integration, globalization, and market openness have contributed to an increased level of integration in the market, which has attracted significant attention in the empirical literature.The volatility of financial markets rises considerably during periods of crisis, especially during the time of the COVID-19 pandemic and the Ukraine-Russia war, which leads to spillover effects across markets.As a matter of course, it is preferable to control and quantify these outbreaks, monitor the progress of already existing crises, as well as deliver a system for "early warning" of emerging crises.
As [1] contend that there has been significant growth in the popularity of cryptocurrency markets over the past few years, and it elevates cryptocurrencies to the status of investment assets.Unsurprisingly, Bitcoin received significant attention since it was the most widespread and first cryptocurrency.As a result of its ability to brute forcing (as it concerns a simple process of trial and error that is repeated quintillions of times every second of the day), Bitcoin allows decentralized systems to issue new coins and verify transactions safely and equitably.There is increasing competition on the Bitcoin network since the increasing Bitcoin price is pushing up miner income and their incentive to mine [2].Approximately 121.36-terawatt hours (TWh) of electricity are consumed each year by Bitcoin, as indicated by Cambridge Bitcoin Electricity Consumption Index. 1 This is more electricity than the actual consumption level of each Argentina's household.An average American family consumes 53 days of electricity, equivalent to a single Bitcoin transaction, according to the Bitcoin Energy Consumption Index published by Digiconomist.It is more likely that the energy market plays a crucial role in the future of cryptocurrencies, as evidenced in these reports.
Several studies have examined the relationship between the gold market and the cryptocurrency market [3,4], the oil market [5][6][7], green commodities and stock markets [8].The interconnectedness within the cryptocurrency market is investigated by [9][10][11].Dutta et al. [12,13] also investigate a nonlinear association between the volatilities of crude oil and precious metals as well as biodiesel feedstock.Gold should be included in a crypto portfolio since it provides a greater degree of diversification [14].Researchers contend that cryptocurrencies are increasingly being regarded as the new gold, which plays an important role in mitigating uncertainty [15,16].Cryptocurrencies have gradually replaced gold as a safe haven for investors during uncertain economic times [17].The gold market has played a modest role during the COVID-19 crisis in limiting risks associated with this crisis.Many scholars, such as Shahzad et al. [18] and Yousaf & Ali [19], provide evidence of the importance of cryptocurrency in hedging, diversifying, and reducing portfolio risk during the COVID-19 pandemic.Since the advent of the cryptocurrency market, the role of the gold market and its interconnections with other markets have been undermined, particularly during COVID- 19.
In 2016 the investment and financial press became highly interested in the role played by Bitcoin.The price of Bitcoin rose by over 1300% in 2017, valuing the total market at over 215 billion dollars.By 2022, it is expected that the entire market value will overreach one trillion dollars.Thus, policymakers and investors should examine and investigate the link between Bitcoin's volatility and returns as well as that of other asset classes.In the event that Bitcoin and other asset classes demonstrate significant returns and volatility spillovers, it might impact asset, risk management decisions, allocation, and selection, as well as regulatory measures which are designed to guarantee the stability of the global financial system.Additionally, this is of interest to politicians who intend to use cryptocurrencies in their foreign exchange reserves or conduct experiments using equivalents of cryptocurrencies.
It is well known that Bitcoin trading consumes significant energy due to its tremendous volume.Accordingly, cryptocurrency is a tool with considerable economic advantages and the potential to accelerate environmental destruction [20].Corbet et al. [21] pointed out that multidimensional advancements in financial technology present a gorgeous picture of current trading while also warning about the risks associated with the future environment.Bitcoin trading has been examined in the current literature in terms of its effect on the sustainability of the environment and the financial market.Jiang et al. [22] have calculated that sustaining the Bitcoin blockchain in 2024 will consume 296.59 terawatt hours, resulting in 130.50 million metric tons of carbon dioxide emissions.An investigation by Polemis and Tsionas [23] examined 50 countries to establish a causal relationship between Bitcoin usage and carbon dioxide emissions.To determine if there is a causal link between carbon dioxide emissions and Bitcoin use, Polemis and Tsionas [23] surveyed data from 50 countries.Incredibly, more down Bitcoin miner returns impact dramatically on conditions of the environment.The stress of this research is on the effect of renewable energy and long-run mining hardware disposals on easing carbon dioxide emissions from Bitcoin on the level of the region.In general, the literature has indicated that prior scholars mainly focused on the link between cryptocurrencies and the energy market [1,24,25].However, no paper explores the nexus between the interlinkages and the volatility arising from these markets.Our paper is the first attempt to fill this gap in the literature.
Our paper makes at least three contributions to the literature.First, as we highlighted previously, our paper is the first work exploring the association between the volatility in the cryptocurrency market and the energy market.Second, this study proposed a methodology for analyzing volatility interlinkages between different types of markets better suited to analyzing these interlinkages.By extending the study of Antonakakis et al. [26] and then Ha [27] and Ha et al. [28], we employ a timevarying parameter vector autoregression (TVP-VAR) with an extended joint connectedness approach to analyze interlinkages between Crypto Volatility and Energy Volatility.Lastly, our research provides a daily dataset from the Crypto Volatility Index (CVI), a decentralized cryptocurrency version of the VIX that lets users insure themselves against market turbulence and temporary loss.Green Bonds (SPGB), Clean Energy (SPGTCLEN), Wind Energy (GWE), Solar Energy (SUNIDX), Natural Gas (NGF), and Crude Oil are used to examine the energy sector's volatility (OVX).The period covered by our time series runs from January 1, 2019, to June 13, 2022.Our attention is mostly paid to the period marked by global uncertainty like COVID-19 or war shock.We compile the S&P Green Bond Index (SPGB), a market-weighted index representing the energy sector's global green bond market.The Macerich Company Global Solar Energy Index Net Total Return (SUNIDX) aims to measure the results of solar energy firms in the world.Moreover, we also have S&P Global Clean Energy Index (SPGTCLEN).Based on an analysis of their products and services, public companies operating in the wind power business are to be measured using the ISE Global Wind Energy Index (GWE).The buyer is required to acquire a particular amount of natural gas under the terms of a Natural Gas Futures contract (NGF).On the delivery of international natural gas businesses, NGF is based.The CBOE crude oil ETF's volatility index (OVX) tracks the volatility anticipated for Crude Oil over the coming 30 days.The fund follows the United States Oil Fund ETF (USO), which primarily owns short-term (1-month) futures contracts on West Texas Intermediate Crude Oil (WTI) on the New York Mercantile Exchange (NYMEX).By using this database, we provide a comprehensive analysis of the link between the volatilities arising from various markets.
In the remainder of this paper, we will organize our discussion as follows.Section "Literature review" provides an overview of the related works on the interconnections between cryptocurrencies and the energy sector.A description of the methodology and data is provided in Section "Statistical analysis and methodology".The empirical results are presented in Section "Results", and we conclude the paper by presenting a summary and policy implications in Section "Conclusions and policy implications".

The cryptocurrency market
In recent years, there has been a growing number of detailed analyses of cryptocurrencies in financial literature.In this regard, an analysis of existing studies in an effort to determine whether cryptocurrencies are considered legitimate investment assets with a legitimate value [21].Another review of the relevant results in previous studies by Kyriazis [29] analyzes how volatility and return spillovers affect the cryptocurrency market.
According to financial studies, empirical results focusing on scrutinizing cryptocurrencies' role as a diversifier, hedges, or safe havens depends on the focused or studied cryptocurrency, time frame, and the assets associated with a set of cryptocurrencies.Therefore, the cryptocurrency market has become a diversifier and/or hedge, a safe haven under market conditions [30].This eventually led to the conclusion that investing in cryptocurrency can be considered a safe move during economic and political crises.Cryptocurrencies, especially Bitcoin, are considered as the New Gold [4,16].Additionally, according to Guesmi et al. [3], investors can reduce risk by including Bitcoin in their portfolios of gold, oil, and stocks.Considering the near future and extreme conditions, Bouri et al. [31] discovered that cryptocurrency has a significant capability to hedge against uncertainty, while uncertainty can be a detrimental factor to the potential profit of Bitcoin.Bouri et al. [31] considered Bitcoin to be an effective diversification tool during economic downturns.Except for gold Kurka [32] failed to find an obvious connection between other traditional assets and Bitcoin.In supporting this view, Smales [33], claimed that Bitcoin returns are unrelated to those of other markets.Additionally, he emphasized how cryptocurrencies should not be viewed as a safe haven until the cryptocurrency market experiences a certain degree of stabilization.Das et al. [34] found that Bitcoin cannot be considered to hold any superiority over other assets.
Cryptocurrencies are now accepted as an accepted asset class while also serving as a hedging instrument and a means for diversification.Therefore, the urge to discover extreme conditions, for instance, the condition that is triggered by the health crisis, is stressed.De la O Gonz alez et al. [35] looked at the behavior of three portfolios made up of equities, bonds, and a cryptocurrency or gold.As a result, while cryptocurrencies show the potential to manage the risk or uncertainty of well-diversified portfolios, few of them were able to complete the task successfully under extreme conditions.Furthermore, despite the stability of gold, it was unable to limit risk as the COVID-19 financial crisis unfolded.Finally, investors should investigate cryptocurrencies in spite of their lower returns in order to achieve greater levels of diversification.Shahzad et al. [18] examined 18 cryptocurrencies while considering both low and highvolatility conditions and discovered significant spillovers during the COVID-19 epidemic.Yousaf & Ali [19], considering three major cryptocurrencies at the same time, concluded investors can benefit from maximum diversification as cryptocurrencies did not fluctuate or witness severe volatility during the pre-COVID-19 period.However, when compared to the COVID-19 period, relationships between different cryptocurrency pairings were intensified.Therefore, the mentioned findings suggest that hedging was more effective during COVID-19, while emphasizing the need for diversification, hedging, and risk control.In this vein [36], scrutinized how the COVID-19 pandemic crisis can have an impact on the cryptocurrency market.Their research pointed out an asymmetric nexus between COVID-19 and the different cryptocurrencies' returns.In addition, they discovered that most cryptocurrencies, including Bitcoin, can operate as a hedging mechanism throughout economic volatility as they cope with the negative impact of COVID-19.Yarovaya et al. [37] gave insights that it is reliant on either bullish or negative market days but does not enhance during COVID-19 considering the herding impact on cryptocurrency markets.By comparing some specific characteristics of the past crisis to the COVID-19 crisis, precious studies also made their way to recommending future research agendas, as contended by Yarovaya et al. [38].Corbet et al. [39] stated how relevant cryptocurrencies could benefit investors in terms of diversification while also proving themselves to be a strong safe haven during the pandemic crisis.While the previous research was concerned with how different currencies intertwined, Corbet et al. [39] were interested in the Chinese financial markets and Bitcoin.Their findings stated that during times of significant economic, and financial upheaval, these assets are considered as contagion amplifiers instead of neither hedges nor safehavens.Under this research, Conlon and McGee [40] noticed how the price of Bitcoin and the S&P 500 both decreased, doing the practice of investing in Bitcoin a safer option during the pandemic downturns becoming unappealing and uncertain.
The situation of COVID-19 is prevailing, therefore, the revision made in this article will concentrate on the first entrance of COVID-19, but also the months in which subsequent waves have been found.Consequently, in the first and second waves of the COVID-19 pandemic, Umar et al. [41] argue that although the trend was most visible in the first waves, slight differences between the first and second waves have been documented [42].Using wavelet coherence analysis, Karamti and Belhassine [43] scrutinized and incorporated financial contagion into fear of the pandemic in connection with US stock markets and international markets in the first and second wave of the pandemic hit the U.S. Regarding the cryptocurrency market, the relationship between Bitcoin and the U.S. COVID-19 fear is stated to be positive under the first wave [43], while in the second wave, the Bitcoin market witnessed the impact of the fear index.This result has ensured investors in the practice of investing in cryptocurrencies as a safe haven.

The cryptocurrency and energy market
Stoll et al. [44] develop a method to estimate Bitcoin mining power consumption by examining IPO offerings of hardware manufacturers, information on mining operations, and the compositions of mining pools.For the purpose of converting electricity consumption estimates into carbon dioxide emissions, the authors make use of the localization of IP addresses.Stoll et al. [44] reveal that Bitcoin consumed 48.2 TWh of electricity, and its annual carbon emissions ranged from 23.6 to 28.8MtCO2.According to Krause and Tolaymat [20], different mining cryptocurrencies consume a distinct amount of energy, respectively.Mora et al. [45] is the first attempt to calculate Bitcoin usage based on the assumption that Bitcoin adoption follows the same rate of adoption as other widely adopted technology.As a result of this new cryptocurrency, it is possible for global warming to exceed 2 C within less than three decades.
Despite the fact that cryptocurrency miners and investors often overlook the ecological impact of networks' energy use, in countries with high electricity costs, cryptocurrency mining profitability has a direct impact on the difficulty and cost of mining cryptocurrency.A number of scholars have also focused their attention on this area in order to establish the value of Bitcoin, identify the relationship between Bitcoin price and mining costs, as well as specifically evaluate the profitability of Bitcoin mining in various regions.Using cointegration models and causality tests, Kristoufek [46] identifies the relationship between the costs associated with mining a Bitcoin and Bitcoin Price Index.The results of this study indicate that the price of Bitcoin influences the cost of Bitcoin mining.There is no surprise in these findings since the growing popularity of this innovative asset, coupled with an increase in price, has attracted many investors and miners to this area, thereby increasing the profitability of mining.
However, Kristoufek [46] notes that after June 2018, Bitcoin mining was only profitable for those professional miners residing in countries where electricity costs are below 0.14 $/kWh.Das and Dutta [47] also report that miner revenues have decreased as a result of the increase in electricity costs and energy consumption.This study highlights a growing problem of mining pools being concentrated in areas with cheap energy, such as China.Therefore, contrary to popular belief that cryptocurrency usage will become more sustainable as a result of renewable energy, this will not be feasible in the near future.Switching to green energy sources and decreasing Bitcoin's carbon footprint may sound like a good idea in theory, but as a result of the high concentration of mining pools in countries that are heavily reliant on coalbased electricity, cryptocurrency networks have an increased carbon footprint.
Bitcoin mining is not necessary for cryptocurrency investments; in addition, some digital currencies are not mineable, and therefore they consume less energy.At first glance, the position appears reasonable, however, research has shown that even small investments in Bitcoin negatively impact portfolio sustainability [48].Consequently, despite the fact that investors may not directly participate in bitcoin mining or construct a portfolio consisting solely of digital currencies, even a small Bitcoin allocation still contributes to an increase in the carbon footprints of their portfolios, even if they do not directly participate in the mining process.Moreover, Bitcoin continues to be the leader of the cryptocurrency market, and its price affects other cryptocurrencies [21].Consequently, portfolios that contain only non-mineable cryptocurrencies are not ethically exempt from the ecological effects of their energy consumption.Thus, any involvement with cryptocurrency entails participation in its CO 2 and adverse ecological impacts.
In the field of cryptocurrency research, numerous studies address specific issues related to the prevalence of cryptocurrency, like its cybercriminal and illegal usage [49].Cryptocurrencies have yet to be extensively studied with respect to their environmental impacts [50].Kristoufek [46] has provided valuable insights regarding the pricing of Bitcoin and the mining costs and revenues, which inspired this paper.This article is also inspired by Baur and Oll [51], that examined how electricity prices relate to Bitcoin energy consumption when viewed in the context of its sustainability and dynamics.In contrast to existing literature, our paper examines the interconnection between the volatility of the cryptocurrency market and the volatility of the energy market.Over the long term, Bitcoin has been demonstrated to generate volatility in fossil fuel and clean energy stocks.As stated by Symitsi and Chalvatzis [52], we further hypothesize that increased volatility in the energy market potentially influences the volatility of the ecological footprint and the energy market regarding both renewable and green energy consumption.

Statistical analysis
Our research provides a daily dataset from the Crypto Volatility Index (CVI), a decentralized cryptocurrency version of the VIX that lets users insure themselves against market turbulence and temporary loss.Green Bonds (SPGB), Clean Energy (SPGTCLEN), Wind Energy (GWE), Solar Energy (SUNIDX), Natural Gas (NGF), and Crude Oil are used to examine the energy sector's volatility (OVX).The period covered by our time series runs from January 1, 2019 to June 13, 2022.We compile the S&P Green Bond Index (SPGB), a marketweighted index representing the energy sector's global green bond market.The Macerich Company Global Solar Energy Index Net Total Return (SUNIDX) aims to measure the results of solar energy firms in the world.Moreover, we also have S&P Global Clean Energy Index (SPGTCLEN).Based on an analysis of their products and services, public companies operating in the wind power business are to be measured using the ISE Global Wind Energy Index (GWE).The buyer is required to acquire a particular amount of natural gas under the terms of a Natural Gas Futures contract (NGF).On the delivery of international natural gas businesses, NGF is based.The CBOE crude oil ETF's volatility index (OVX) tracks the volatility anticipated for Crude Oil over the coming 30 days.The fund follows the United States Oil Fund ETF (USO), which primarily owns short-term (1-month) futures contracts on West Texas Intermediate Crude Oil (WTI) on the New York Mercantile Exchange (NYMEX).Regarding the cleaning process, we removed all zero and missing observations from our daily collected data as required by our empirical approach.We also checked for outliers and removed them if they presented.We use the first log-differenced series.Specific numbers may be used to calculate the growth rate of these indicators because non-stationary systems are more likely to be produced by our indications of study.
These systems are built on the unit-root test statistics Elliott et al. [53] introduced.
The average return for all series in Table 1 is positive.In particular, the Crypto Volatility Index and Crude Oil proved to be the riskiest assets in the data of Panel A because of their very high variation.Significantly, it is discovered that all series are clearly leptokurtic.These results imply that their distributions have a bigger tail than the conventional distributions.According to Jarque & Bera [54], all indicators are not regularly distributed.The ERS unit root test of Elliott et al. [53] finds that all indicator returns are stationary at a 1% significance level.Additionally, the weighted portmanteau test [55] demonstrates that autocorrelation is present in both the returns and squared returns.These outcomes back up the use of our method.We use a TVP-VAR technique with a time-varying variance-covariance structure to represent the interconnectedness of the series.Our article is the first to attempt to understand how the volatility of crypto is impacted by the energy market, particularly during the COVID-19 pandemic.We thus look into the relationship between several indicators that altered prior to and during the COVID-19 epidemic.
A comprehensive summary of these indicators in these two subsets is also provided  Significantly, all variances increase once COVID-19 arrives, increasing the volatility of all other indicators save the Crypto Volatility Index.In other words, Figure 1 illustrates that Natural Gas is viewed as a safe haven for asset managers because of their improved earnings throughout the crisis.This is because Natural Gas encounters economic, political, and extraordinary occurrences like the COVID-19 epidemic.The findings of the weighted portmanteau test and the ERS unit root test on these indicators across these two periods are more likely to be identical to those discovered through testing on the complete sample.These findings suggest that the optimal way to simulate the interconnectedness of the series is a TVP-VAR approach with a timevarying variance-covariance structure.

Empirical methodology
In this part, we follow Antonakakis et al. [26] and then Ha [27] to extend the TVP-VAR connectedness technique, which was developed by Diebold and Yilmaz [56].The Bayesian information criterion (BIC) suggests that the TVP-VAR model be estimated with a lag length of order one in our article: where A t and R t are P Â P dimensional matrices, whereas y t , y tÀ1 and ѱ t are pP Â 1 dimensional vectors.R t is a P 2 Â P 2 dimensional matrix, whereas vec A t ð Þ and u t are P 2 Â 1 dimensional vectors.This method includes all indices A t ð Þ changing throughout time, as well as the connection between series.Moreover, the R t and R t variance-covariance matrices are considered to be time-varying.Almost all prior research has proven that variances and covariances change over time, particularly in the financial market; this shows the altering market and risk ratio.
This article turns TVP-VAR into a TVP-VMA model in the next step: y t ¼ P 1 h¼0 N h, t ѱ tÀ1 where N 0 ¼ I Z .We assume that ѱ t is a vector of shocks (symmetric but not orthogonal) with P Â P covariance matrix E ѱ t ѱ 0 t À Á ¼ R t that can change over time.As a result, the ɲ-step estimate error is expressed as: X t fi ð Þ ¼ y tþɲ À E y tþɲ y t , y tÀ1 , . . .ð Þ ¼ P ɲÀ1 l¼0 N l, t ѱ tþɲÀl with a forecast error covariance matrix equal to: The suggested approach is based on [57] and [58] ɲ-step forward generalized forecast error variance decomposition (GFEVD).The (scaled) GFEVD, q˜ʓ ij, t , can be read as the impact of a shock in indicator j on indicator i and is written as: where e i is a pP Â 1 zero selection vector with unity on its ith location and X gen ij, t ðɲÞ is the decreased level of indicator i's ɲ-step prediction error variance owing to controlling the unexpected shocks of indicator j Diebold and Yilmaz [56] suggested standardizing the P P j¼1 X gen ij, t ðɲÞ 6 ¼1 to unity using the row sum, leading to the generalized spillover panel, gST ij, t : Several spillover summary estimates can be derived from the generalized spillover table, including the total directional connectedness from others to indicator i.The total directional connectedness index from a shock in indicator i to others is expressed as follows: Indicator I's net total directional connectedness indicates whether it impacts the system more than it is impacted by it, among the core metrics of the connectedness approach: , indicator i is a net receiver or transmitter of shocks, meaning that indicator i is affected by or affecting the system.The TCI is meant to be the average total directional connectedness from (to) others, and it is calculated as follows: Lastly, the connectedness method gives evidence of the pairwise interrelationships of two indicators through the idea of net pairwise directional spillovers, which are described as: The major purpose is to determine the q˜ʓ ij, t equivalence for the mutual connectedness method, called j˜ʓ ij, t , that meets these criteria: To do this, we must adopt the technique of [59].As a result, the recommended computation of Equation ( 12) must be correct.As the row total of the original and joint connectedness tables must equal 1, the joint connectedness table's diagonal components must also remain the same.As a result, the scaling factor varies per row, yielding the given formula: The sole difference between our g soaring and the one that arises from the joint connectedness technique is that our method allows greater flexibility because each row has its own soaring element.Then, the steps below must be arranged: Furthermore, by varying the soaring parameter by row, the net total and pairwise directional connectedness metrics may be calculated based on: Our approach is expected to provide more precise findings than methods employed in previous studies because they solve the drawbacks of the row sum normalization technique, although the explanations are equal to those of the original connectivity approaches by Caloia et al. [60].

Results
The next section displays the average and dynamic findings for the connectedness measures used in our study.The total connectedness index (TCI) average value is based on the entire sample of data.TCI is initially introduced, then a dynamic evolution of the TCI through time is shown.The latter strategy is essential for comprehending how the TCI reacts to different economic situations.Over the course of our research period, political trends are also observed.We also assess data for total net connectedness and net pairwise connectedness inside our proposed system.This connection helps us better comprehend the market for seven indicators, including the Crypto Volatility Index, Green Bonds, Clean Energy, Wind Energy, Solar Energy, Natural Gas, and Crude Oil.It is essential to remember that each indicator has the potential to function as either a net shock transmitter or receiver.Finally, we use the joint spillover index developed by Lastrapes and Wiesen [59].These findings may be utilized to look into the causes of changes in the link of networks between different metrics.

Variation in average dynamic connectedness over time
Using the whole set of data from January 1, 2019, to June 13, 2022, Table 2 displays the average results for the interlinkages of various indicators inside the network of various indicators.This table's diagonal portion summarizes a single indicator's variation, which is determined by its own shocks.While the off-diagonal components explain how this indicator affects other indicators' fluctuation (FROM) and how other indicators affect the fluctuation of this indicator (TO).In particular, Table 2 shows the impact of each individual indicator on the prediction error variance of each indication, whereas the columns show the independent impact of each type of indicator on each other.
For the whole collection of data, the TCI average value is 44.06%.It is demonstrated that modifications to this network may account for 44.06% of the variation in the network of indicators under study.This suggests that idiosyncratic factors are responsible for roughly 55% of the error variance in the system.The final row of Table 2 shows each indicator's contribution.According to this research, crude oil and clean energy substantially transmit shocks and volatility to other system indicators.
According to Tiwari et al. [61], clean energy leads all other markets and is thought to be the primary net shock transmitter in the overall network.It's amazing that the network's Crypto Volatility Index is susceptible to shocks.We discover that Solar Energy is receiving the biggest shock from the Crypto Volatility Index, amounting to 1.33 percent.Additionally, the volatile renewable energy market includes Clean Energy, Wind Energy, Solar Energy, and Green Bonds.They experience shocks from fossil fuels like Natural Gas, Crude Oil, and the Crypto Volatility Index.In decreasing order, the CVI is also affected adversely by shocks from other energy indicators, especially Wind Energy, Clean Energy, Solar Energy, Crude Oil, Green bonds, and Natural Gas.The same is true for clean energy, solar energy, and green bonds, all of which are net recipients of numerous shocks.Solar Energy and Green Bonds are the most significant net receivers when network metrics fluctuate.
By splitting the sections of observations by the COVID-19 crisis era, this study focuses on the premise that each indicator plays a unique function during different periods.The history of the system prior to the COVID-19 breakout can only be partially explained by the system of all Similarly, during the COVID-19 crisis, idiosyncratic effects can be responsible for around 54% of the system's forecast inaccuracy variation.These results provide credence to the idea that these markers frequently move in lockstep, particularly under unclear circumstances like the COVID-19 epidemic.The entire time, Crypto Volatility Index handles two completely different tasks.CVI is more likely to be a network shock transmitter at the normal time.In this uncertain circumstance, CVI appears to be a net receiver.It means that the cryptocurrency market is particularly vulnerable to and greatly influenced by outside factors.Fossil fuels have been found to shock green energy more than the Crypto Volatility Index.Prior to the COVID-19 pandemic, wind energy only transmitted a small number of shocks to CVI; however, following the COVID-19 epidemic, wind energy has become the main transmitter of shocks to CVI.We can demonstrate empirically that the Crypto Volatility Index contributes to the explanation of the energy market volatility.

Dynamic total connectedness
It's important to note that those typical results are most useful as a summary of the underlying interconnection.Average results are constrained in the COVID-19 epidemic wave to enable the analysis of interconnection across a network of determinants.
A more dynamic framework of analysis is therefore needed.It considers not only how the TCI has changed over time but also how the role played by certain indicators within the research network may have changed.For instance, changes from net sending to net receiving must be considered.Figure 2 of the study's dynamic total connectedness data illustrates the TCI's intertemporal evolution.The TCI values change considerably during the course of our study period.The COVID-19 outbreak causes the TCI values to peak at about 60% by the end of 2019.Particularly, the more significant the TCI values, the more interconnected the indicators are.The TCI values' amazing stability at approximately 40% is astounding.When there is a shock from the COVID-19 outbreak, the TCI levels peak at more than 60%.Previous studies, such as those by [62] and [63], have demonstrated that the interconnection of diverse commodities markets rises in uncertain periods, such as the global financial crisis (2007)(2008)(2009).The value of the TCI then starts to decline around the end of 2021.The lowest percentage is close to 10%.Our findings and those of earlier research demonstrate that the TCI's dynamic development responds to COVID-19 shocks.As uncertainty rises, the link deepens.Finally, the Diebold and Yilmaz [56,64] technique may be used to verify all of the peaks and troughs indicated above.

Net total and pairwise directional connectedness
The findings of net connectedness are then examined.These findings categorize various market categories as either net transmitters or net receivers.The classification described in Section "Variation in average dynamic connectedness over time" is contrasted with the current dynamic technique, which enables us to identify potential changes between the two roles under examination.To put it another way, the function of indicators will vary between net shock emitters and receivers in the renewable energy industry depending on the research period and particular types of indicators.We refer to total net connection in our essay.It is indicated that an indicator functions consistently across time in regard to all other indications.The results of our study on pairwise net connectedness, which entails examining pairs of indicator types, are described in the following sections.Their objectives are to examine how their bond has evolved through time with respect to several potential roles.These outcomes are displayed in Figure 3.The positive and negative numbers represent the respective indicators' net transmitting and receiving functions.In 2019, the cryptocurrency volatility index appeared to be a net transmitter.A 51% attack on the blockchain and government backing for the Ethereum community's creativity are two further reasons why 2019 is the year of cryptocurrencies.Due to the Russia-Ukraine Crisis, this indicator shifted from being a net receiver of shocks in 2020 to late 2022 to being a net transmitter of shocks starting at the end of February 2022.When crises like the COVID-19 outbreak develop, it shows that the cryptocurrency industry is an extremely dangerous investment.It is concerning that the Russia-Ukraine Crisis has increased uncertainty over the European gas supply's security and price volatility.Several cryptocurrencies also experienced collapses in 2022 during the same period.Due to this, the cryptocurrency market as well as the energy industry, saw market instability.The OVX and SPGTCLEN indicators are net shock transmitters in both the time leading up to and following the health crisis.In stark contrast to OVX's trajectory, SUNIDX and SPGB follow different paths.GWE and NGF, on the other hand, plan to eventually fill both positions.To sum up, during severe crises, both clean energy, and crude oil are net long-term shock transmitters inside our network.However, a sizable portion of the variation in the energy market is accounted for by the Crypto Volatility Index, and its dominance is just transitory.
It is important to note that the original method's normalizing process lacks any theoretical underpinnings and hence represents an unexpected manner of connectedness normalization.Therefore, utilizing the theoretically derived measurements from [59] is better.Next, we concentrate on the net pairwise connectedness findings displayed in Figure 4. Our goals are to identify the critical function of the Crypto Volatility Index within our system of numerous indicators and to demonstrate how the renewable energy sector fluctuates.Our research initially examines the effects of spillovers related to cryptocurrency volatility.Before 2020, all energy market indicators, such as clean energy, solar energy, green bonds, crude oil, and natural gas, are subject to shocks from the cryptocurrency volatility index.In other words, the renewable energy industry depends on the cryptocurrency market's volatility up until 2020.The cryptocurrency market prospered in 2019, and the energy market volatility has a lot to do with the crypto market volatility.The Crypto Volatility Index also contributes to the identification of the volatility of SPGB, SPGTCLEN, GWE, SUNIDX, and NGF, especially during the COVID-19 epidemic.The COVID-19 epidemic and the beginning of 2020 saw the Crypto Volatility Index as the main supplier of shocks for the renewable energy industry, but by late 2020 the crypto market was the recipient of shocks from clean energy, wind energy, and green bonds.It is important to note that, starting in 2021, the network with renewable energy fulfills both functions according to the Crypto Volatility Index.The crypto market was shaken out of its pricing slumber by many rounds of unrestricted central bank stimulus spending, which also improved the economic case for cryptocurrencies as a store of wealth.The cryptocurrency market is a hedging asset from the energy industry as of early 2020.However, interest in smart grids, renewable energy, and electric cars has increased.Authorities have also announced plans to phase out new gasoline-powered cars from over the course of the following 10 to 15 years, from the United Kingdom to California.Therefore, when energy demand declines as a result of the COVID-19 health crisis, the volatility of cryptocurrencies speeds up the switch from nonrenewable to renewable energy use.When situations like COVID-19 arise, renewable energy stocks become a desirable investment avenue to draw investors.Clean energy, wind energy, and solar energy are significant transmitters since they have sparked the clean energy market's growth expectations.Scholarly, the Crypto Volatility Index is a significant factor in influencing the market volatility for green bonds.It demonstrates that, particularly in times of crisis, Green Bonds are unable to mitigate the risk that the cryptocurrency market poses to the renewable energy industry.Green bonds are viewed as a way to deal with the energy market's volatility when COVID-19 appears, given the energy market correlations.Notably, Crude Oil may eventually act as a transmitter of shocks to the Crypto Volatility Index.In terms of scale, Crude Oil spillover activity has been rather substantial since late 2020, but from the end of 2021 to the conclusion of our data, it has been dropping.The crude oil market is frequently erratic, particularly during times of crisis.As a result, crude oil is now the primary shock transmitter in the network.We can see the outcomes more clearly in Figure 4.The significance of the Crypto Volatility Index on the energy market is seen in Figure 5.

Robustness checks
In the literature, scholars may concern that the Crypto Volatility Index is built utilization of information from crypto derivative trading platforms such as Deribit.Deribit is a nonregulated platform with a section on its website that features the March 2020 price crash [65].Some items listed here include deviations between the mark and actual spot price of the derivative contracts and an introduction of circuit breakers to shut down the platform under stress, which seemingly triggered at least twice in this period.Furthermore, unregulated platforms can be known to be prone to market manipulation [66].Accordingly, it is not clear how this influences CVI, thus, processing these in a calculation may not properly safeguard against producing inconsistent and extreme outcomes.
We used the investor sentiment index developed by Bouteska et al. (2022) for a robustness check and replicating our model with a new dataset.This index is used as a proxy for investor sentiment toward the cryptocurrency market and Bitcoin prices.Bouteska et al. [67] contend that the investor sentiment composite index is best at predicting the future directions of the Bitcoin market.The results are reported in Appendix.In general, all our conclusions still remain.

Conclusions and policy implications
Our study applies an extended TVP-VAR framework to measure the network connectedness of seven indicators, including the Crypto Volatility Index, Green Bonds, Clean Energy, Wind Energy, Solar Energy, Natural Gas, and Crude Oil, in a time-varying manner.We also use the strategy Balcilar et al. [68] proposed, which offers greater flexibility and lets us achieve the net pairwise connectedness measurements.The Crypto Volatility Index, Green Bonds, Clean Energy, Wind Energy, Solar Energy, Natural Gas, and Crude Oil are among the daily datasets we gathered for this study.The period covered by our time series runs from January 1, 2019 to June 13, 2022.
Our findings demonstrate that all the investigated indicators are just marginally related when considering the entire set of data.This research specifically demonstrates the existence of a timevariant of system-wide interlinkages driven by the COVID-19 outbreak and the Russia-Ukraine Crisis.Additionally, we provide empirical proof that the Crypto Volatility Index, wind energy, and natural gas may have had a net transmitter or receiver effect during our sample.While the system pushes Solar Energy and Green Bonds during our sample, Crude Oil and Clean Energy dominate the system.While net total directional connectedness over quantile reveals that natural gas and crude oil are net shock transmitters throughout the 2020-2022 timeframe, the Crypto Volatility Index, clean energy, solar energy, and green bonds steadily receive all other indicators.According to pairwise connectedness, the Crypto Volatility Index appears to be a net transmitter of shocks to all energy market indicators prior to the COVID-19 epidemic and transforms into a net receiver of shocks from renewable energy sources like clean energy, wind energy, and green bonds by the end of 2020.Most of the spillover effects from the Crypto Volatility Index are experienced by green bonds.The primary network shock source for the Crypto Volatility Index is crude oil.

Theoretical implications
In order to assess the effects of significant events like the COVID-19 epidemic and the Russia-Ukraine Crisis on the dynamic connectedness between these indicators, we are the first to thoroughly discuss the connectedness between these indicators, namely the Crypto Volatility and Energy Volatility, in our article.Using this unique technique, we can determine the net pairwise connectedness-a measure of transmission mechanisms between Crypto Volatility and Energy Volatility.This study is anticipated to offer important insights and cautions regarding the contagious effects of uncertain events and policies to both investors and regulators.

Practical implications
Our findings have important policy ramifications for investors and authorities, as well as recommendations based on the interactions between the various factors and their spillover effects.Policymakers can design the most effective policies to lessen the vulnerabilities of these indicators and reduce the spread of risk or uncertainty across them by having insightful knowledge about the primary antecedents of the contagions among these indicators, namely the Crypto Volatility Index, Green Bonds, Clean Energy, Wind Energy, Solar Energy, Natural Gas, and Crude Oil.Our findings highlight the potential danger of either low or excessive diversification for authorities in these metrics by demonstrating the significant connections between seven indicators.Our findings highlight the growing connections between unforeseen and extremely unpredictable occurrences, such as the COVID-19 outbreak and Russia-Ukraine.According to our research, a shock to one typical indicator affects the entire network, suggesting that the cryptocurrency market encourages the growth of renewable energy.Additionally, as energy demand declines as a result of the COVID-19 health crisis, the volatility of cryptocurrencies speeds up the switch from nonrenewable to renewable energy use.Cryptocurrency networks have an increased carbon footprint due to the high concentration of mining pools in countries where coal is a major source of electricity.After China banned miners in the Spring of 2021, the carbon footprint of Bitcoin increased as miners relocated to areas with a higher share of fossil fuels in the energy mix.Increasingly, it has become apparent that one of the major factors driving the switch to renewable energy solutions was their relative cost-effectiveness in comparison with fossil fuels.The mining industry is becoming more aware of the fact that it has a unique opportunity to score multiple wins by switching to renewable energy solutions.Firstly, renewable energy is more cost-effective in the long run than fossil fuels.Renewable energy also provides a more fixed cost than fossil fuels, which can suffer from periods of high and sustained volatility, such as the world is currently experiencing.Notably, fossil fuels such as coal, which are largely used in emerging markets, are difficult to replace with alternative energy sources.Hydrogen-based technologies provide a pathway for greening the steel production process, however, the incentives are currently weak due to a lack of carbon pricing.Second, fossil fuel power plants have a design lifespan of at least 30 to 40 years.There is no doubt that coal plants will remain in use for some time to come unless there is a dramatic reduction in the cost of renewable energy or policy makers intervene.Third, moving away from coal typically means losses for the domestic mining industry and its workers.Strong domestic mining interests may complicate and delay the phaseout of coal in major coal-consuming countries such as China and India.The rapid transition from coal to natural gas in the United States resulted in a decline in coal mine employment, a record number of bankruptcy filings among coal mining companies, and a sharp decline in coal mining stocks.A similar transition in some coal-producing countries could imperil financial stability, as banks take losses on investments in obsolete mines and power plants-socalled "stranded assets." In addition, policymakers may find the conclusions of this article helpful in their attempts to improve public welfare, which is directly impacted by crypto-and energy-market volatility.Useful significant insights include the fact that risk and uncertainty in the cryptocurrency market have an influence on the energy industry and vice versa.Therefore, it is essential to include them while developing policies for a vulnerable minority in order to improve society's welfare.

Limitations and directions for future research
Three restrictions should be taken into consideration while interpreting the study's results.First and foremost, it is crucial to stress that there is no one-size-fits-all rule or general pattern governing how risk occurrences affect both net, or pairwise spillovers.Second, from the perspective of indicator integration, spillover size is important.A certain market system will be significantly impacted by changes and shocks resulting from other indicators if the spillover is substantial.To mitigate the detrimental effects of external shocks, the government must adopt a number of actions.Authorities should pay attention to risk sources that are frequency-specific.More emphasis should be placed on minimizing the detrimental impacts of shortterm return spillover and long-term volatility spillover in the coordination of international regulatory policies relating to various indicators.Last but not least, measuring the portfolio advantages of diversity is a significant expansion, given that many academics take into account the spillover impact across many measures.We put it on the back burner in the meantime.

Contributions
Nguyen Thi Thanh Huyen, Nguyen Hong Yen and Le Thanh Ha contributed to all stages of preparing, drafting, writing and revising this review article.Nguyen Hong Yen collected the data and data analysis.Le Thanh Ha made a substantial, direct, and intellectual contribution to the work during different preparation stages.Nguyen Thi Thanh Huyen read, revised and approved the final version of this manuscript.

Figure 1 .
Figure 1.Crypto volatility index, green bond, clean energy, wind energy, solar energy, natural gas, and crude oil returns.

Figure 4 .
Figure 4. Dynamic net pairwise directional connectedness: Other indicators to Crypto Volatility Index.Panel A: Whole sample Panel B: During the COVID-19 health crisis

Figure 5 .
Figure 5. Dynamic net pairwise directional connectedness: Changes in Crypto Volatility Index to other indicators.Panel A: Whole sample Panel B: During the COVID-19 health crisis Disclosure of potential conflicts of interestResearch involving Human Participants and/or Animals Informed consent

Figure A. 3 :
Figure A.3: Dynamic net pairwise directional connectedness: Green energy volatility to Investor Sentiment.

Figure A. 4 :
Figure A.4: Dynamic net pairwise directional connectedness: Investor Sentiment to green energy volatility.
during-COVID-19 (from January 1, 2020, to June 13, 2022).The general changes between the two eras of the chosen indicators are shown in

Table 1 .
Most of our included indicators, with the exception of SPGB, had larger average returns during the COVID-19 period than during the pre-COVID-19 period.However, the pre-COVID-19 period had negative average returns for NGF and OVX.The mean return of the global Crypto Volatility Index is also greater after the COVID-19 pandemic catastrophe, which shook the financial system, occurred.Solar energy has the highest mean return throughout the COVID-19 period of all renewable energy sources.It demonstrates that this signal turns into the most alluring.It is amazing that COVID-19 has caused the average returns on natural gas (NGF) and crude oil (OVX), two fossil fuels, to move from negative to positive.Investors find the fossil energy sector more alluring during times of crisis.

Table 2 .
Averaged joint connectedness between CVI and green energy.