The Differential Influence of Social Media Sentiment on Cryptocurrency Returns and Volatility Throughout COVID-19 > 자유게시판

• 목록
• 답변
• 글쓰기
• 게시판 리스트 옵션
  • 수정
  • 삭제

Received 2021 Aug 16; Revised 2022 Jul 27; Accepted 2022 Sep 25; Issue date 2023 Jun.

Since January 2020 Elsevier has created a COVID-19 useful resource centre with free info in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the corporate's public news and knowledge website. Elsevier hereby grants permission to make all its COVID-19-related research that is out there on the COVID-19 useful resource centre - including this research content material - immediately out there in PubMed Central and different publicly funded repositories, such because the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the unique supply. These permissions are granted free of charge by Elsevier for as long because the COVID-19 useful resource centre stays lively.

This analysis investigates the consequences of a number of measures of Twitter-based mostly sentiment on cryptocurrencies throughout the COVID-19 pandemic. Innovative financial, in addition to market uncertainty measures primarily based on Tweets, alongside the traces of Baker et al. (2021), are employed in an try and measure how investor sentiment influences the returns and volatility of main cryptocurrencies, growing on non-linear Granger causality assessments. Evidence means that Twitter-derived sentiment primarily influences Litecoin, Ethereum, Cardano and Ethereum Classic when contemplating imply estimates. Moreover, uncertainty measures non-linearly affect each cryptocurrency examined, at all quantiles aside from Cardano at lower quantiles, and each Ripple and Stellar at each lower and better quantiles. Cryptocurrencies with decrease values are found to be unaffected by investor sentiment at extreme values, nonetheless, show to be worthwhile resulting from more aligned investor behaviour.

Keywords: Sentiment, Uncertainty, Cryptocurrencies, COVID-19, Pandemics, Black Swans

Cryptocurrencies have generated much debate surrounding whether the disintermediation and online technological progression of finance at a worldwide level merits the levels of danger which were generated through further components similar to cybercriminality, regulatory ambiguity, fraud, manipulation, and broad propensity to generate irrational exuberance (Akyildirim et al., 20 Akyildirim et al., 2021, Gandal et al., 2021). These matters have drawn even bigger opprobrium because of the development of the COVID-19 pandemic that began as a well being disaster and rapidly developed into a world financial crisis. This pandemic has generated substantial global financial market volatility and has driven investors to seek alternative assets to preserve their portfolios at a passable risk-return commerce-off degree. Cryptocurrencies have attracted speculators as well as technology-fluent buyers (Lee et al., 2020), some of which have been attracted by the product validation offered by a number of extensively identified public figures and firms, a lot of whom have had restricted, if not no experience at all growing and promoting technologically developed products (Akyildirim et al., 2020, Cioroianu et al., 2021, Fletcher et al., 2021). As an asset, cryptocurrencies have been thought of to be primarily employed for hypothesis purposes but not instead currency and medium of alternate (Fry, 2018, Kyriazis et al., 2020).

Investigating the determinants of the returns and volatility of cryptocurrencies has been the main target of a broad variety of tutorial research (Dyhrberg et al., 2018, Eross et al., 2019, Katsiampa, 2017, Katsiampa et al., 2019a, Katsiampa et al., 2019b, Akyildirim et al., 2020 Akyildirim et al., 2021, Papadamou et al., 2021, Sensoy et al., 2021). This analysis sets out to build on this work, and additional examine whether non-linear causal linkages exist between Twitter-derived measures of economic and market uncertainty and the largest cryptocurrencies throughout the COVID-19 pandemic. These Twitter-based measures of investor sentiment developed by Baker et al. (2021), advance the already-present and extremely standard Economic Policy Uncertainty (EPU) indices. To assemble the TEU-USA indicator, the authors tokenize and use the lower-case variations of all tweets in their sample, whereas counting the frequency of tweets that contain keywords associated to the economy1 . Further, duplicate tweets are removed, and solely people who combine economic and uncertainty terms remain within the sample after the Random Forecast Classifier is used to tell apart whether the US is the placement of the tweets or not. The impacts of these modern measures of investor sentiment on cryptocurrency imply and volatility are investigated by employing the extremely-refined non-linear quantile causality methodology of Diks and Panchenko (2006). This is taken into account to be more superior than the standard Granger causality methodology and can distinguish non-linear impacts which are removed from simply discernible when conventional methodologies are employed. Econometric estimations are undertaken to cowl the complete interval of the COVID-19 pandemic from its beginnings in January 2020 to the current. This examine is related to previous educational work investigating the impacts of uncertainty measures on cryptocurrencies (Fang et al., 2020, Wang et al., 2020) and the linkages of COVID-19 with financial markets (Conlon et al., 2020, Raheem, 2021, Sarkodie et al., 2021).

This study is the primary to scrutinise the consequences of such a large range of revolutionary uncertainty indices upon the returns and volatilities of cryptocurrencies, specifically shedding gentle upon the nexus between Twitter-based mostly investor sentiment, digital forms of liquidity and investment, and the COVID-19 pandemic, providing investors with beneficial insights to help the event of extra correct determination-making via adversarial situations. Presented outcomes indicate that Bitcoin, Ethereum, Bitcoin Cash, and Litecoin are non-linearly influenced in the mean by the selected Twitter-derived financial uncertainty indices. However, solely Ethereum, Bitcoin Cash, and Cardano are found to be non-linearly brought about in the mean by Twitter-based market uncertainty measures. The majority of analysed cryptocurrencies are discovered to be receivers of uncertainty affect at all quantiles investigated, nevertheless, those with low nominal values appear to stay unaffected by Twitter-derived sentiment indicators. Such a phenomenon could be very useful for hedging functions in portfolios that encompass fashionable or traditional financial belongings, where some cryptocurrencies can serve as safe-havens throughout turbulent financial and market situations. Such outcomes are discovered to be robust when contemplating different methodological specifications.

The remainder of this paper is structured as follows. Section 2 supplies a concise overview of the literature, together with a proof surrounding the connection between EPU uncertainty measure’s interactions with financial markets. Section three presents an overview of the info and an explanation of the methodology which has been adopted for estimation. Section 4 presents an outline of the results, whereas Section 5 concludes and suggests avenues for additional research.

A significant number of educational papers have focused on the investigation of investor sentiment within the constructions of fashionable financial belongings. This examine contributes to 4 specific strands of educational analysis surrounding this area. First, we give attention to the influence of financial policy uncertainty on cryptocurrencies. Specifically, Wang et al. (2020) provide proof that larger ranges of EPU lead to larger cryptocurrency returns. US EPU is revealed to result in elevated Bitcoin volatility and trading volume after EPU spikes whereas UK EPU is discovered to not be as influential. Moreover, direct spillovers from US EPU to UK EPU are detected. Beneki et al. (2019), argued that financial policy uncertainty strengthens the connection between Bitcoin and Ethereum and gives weak diversification advantages concerning portfolio efficiency. In a somewhat completely different perspective, Fang et al. (2020) word a connection between news-primarily based implied volatility and a direct affect upon the long-time period volatility of five major cryptocurrencies in each a unfavorable and vital manner. This result is also discovered to be legitimate when the global Economic Policy Uncertainty index is considered. Investor sentiment is revealed to be more essential than financial fundamentals to foretell cryptocurrency volatility. Apart from strictly specializing in economic policy uncertainty, studies associated to geopolitical uncertainty effects or commerce uncertainty impacts on cryptocurrencies have had a central function in earlier literature growth. Gozgor et al. (2019) establish a significant relationship that runs from US Trade Policy Uncertainty (TPI) to Bitcoin returns and presents further evidence of regime changes through the durations between 2010-11 and 2017-18. This connection is revealed to be highly effective throughout regime changes. Further, Baker et al. (2021) make the extension of their Economic Policy Uncertainty index by focusing on tweets to derive investor sentiment. They create 4 progressive financial sentiment indices and 4 market sentiment indices which might be all based mostly on tweets, arguing that Twitter customers bear massive similarities with journalist perceptions about danger and uncertainty. Moreover, Lehrer et al. (2021) perform an out-of-sample exercise and reveal that when social media sentiment is included then the forecast accuracy of a preferred volatility index may be improved, especially in the brief run. High-frequency information are discovered to be favourable for forecasting. Further, Tumasjan et al. (2021) argue that there is a positive linkage between signalling and enterprise capital valuation, however Twitter sentiment just isn't discovered to be related to funding success in the long term.

A collection of related papers that concentrate on impacts of Twitter-primarily based investor sentiment on cryptocurrencies includes that of Philippas et al. (2019), who study whether or not Bitcoin value jumps are associated with Twitter and Google Trends informative indicators. Outcomes by the twin diffusion mannequin adopted reveal that Bitcoin market values are partially led by the momentum of media attention in social networks, and investors ask for data to make investing decisions. Similarly, Li et al. (2021) identified that bi-directional causalities and spillovers exist among nearly all of the twenty-seven cryptocurrencies investigated and investor attention. It is underlined that when investor sentiment is predicated on a mix of Twitter and Google search knowledge, these interlinkages are more apparent. Huynh (2021) assessed the impact of President Trump’s tweets on Bitcoin value and buying and selling volume, arguing that destructive sentiment tweets prove to be significantly extra highly effective than constructive ones as considerations the predictive energy about returns, trading volume, realised volatility and jumps in Bitcoin markets. Kraaijeveld and De Smedt (2020) focus on the predictive powers of Twitter sentiment and adopt a lexicon-based mostly sentiment analysis and bilateral Granger causality for finding out the nine largest cryptocurrencies. It's revealed that Twitter considerably affects the returns of Bitcoin, Bitcoin Cash and Litecoin, whereas also EOS and TRON if a bullishness ratio is employed. Further, Wu et al. (2021) adopted the Twitter-based mostly EPU and TMU measures and reveal that the Twitter-derived financial uncertainty significantly influences the Bitcoin, Ethereum, and Ripple values expressed in US dollars. These impacts are found to be constructive. Moreover, Naeem et al. (2021) centre their interest on the FEARS index and the Twitter sentiment index and argue that the happiness sentiment is a stronger predictor of cryptocurrency returns. Predictability is discovered to be pushed largely by social media sentiment moderately than macroeconomic information. Furthermore, Umar et al. (2021) employ a range of sentiment indicators to seek out which better expresses the bubble phenomenon of GameStop. Media-pushed sentiment indicators reveal the massive ranges of inefficiency that buyers may create in markets2 . Social media platforms might be helpful for monitoring such investing behaviour. Finally, Béjaoui et al. (2021) argue that there's a strong dynamic nexus between Bitcoin and social media that additionally holds throughout the COVID-19 era, both social media and Bitcoin costs are revealed to be influenced by a substantial extent of this health crisis.

Conlon et al. (2020) support that Bitcoin and Ethereum don't represent a secure haven regarding nearly all of worldwide fairness indices investigated and when these digital currencies are included in portfolios they end in larger draw back risk. However, Tether is revealed to exhibit some protected-haven traits in the direction of several examined indices3 . Guo et al. (2021) discover that the COVID-19 disaster has led to a stronger contagion impact between Bitcoin and developed markets. It's identified that both US and European markets stay contagion sources to Bitcoin whereas gold, the US dollar and bond markets are recognized as receivers of contagion results. The pandemic can also be discovered to have weakened the diversifying, hedging or secure haven properties of Bitcoin. Goodell and Goutte (2021) recognized that the COVID-19 pandemic positively influences Bitcoin market values. This phenomenon is extra apparent after the March 2020 crash in financial markets. Furthermore, Huang et al. (2021), via the application of a Bayesian Panel-VAR methodology reveal that main economies take pleasure in diversification benefits and danger mitigation within and throughout borders attributable to Bitcoin in the course of the COVID-19 era4 .

3. Data & methodology

Estimations happen in the course of the interval since the beginning of the COVID-19 pandemic, designated as the official WHO announcement identifying the existence of a world pandemic on 1 January 2020, by the interval ending 25 July 2021. To conduct econometric estimations to detect causality, eight Twitter-derived uncertainty measures have been used5 . These indices are primarily based on the work of Baker et al. (2021), and consist of four financial uncertainty indices (TEU-ENG, TEU-USA, TEU-WGT, and TEU-SCA) and 4 market uncertainty indices (TMU-ENG, TMU-USA, TMU-WGT, and TMU-SCA), the place TEU stands for Twitter Economic Uncertainty and TMU represents Twitter Market Uncertainty6 . Moreover, every day information based on the ten largest cryptocurrencies by market capitalisation during the examined period are adopted for our empirical estimations7 . More particularly, the most important cryptocurrencies by capitalisation examined are Bitcoin (BTC), Ethereum (ETH), Binance Coin (BNB), Cardano (ADA), Ripple (XRP), Dogecoin (DOGE), Bitcoin Cash (BCH), Litecoin (LTC), Ethereum Classic (Etc), and Stellar (XLM).

Table 1 illustrates the descriptive statistics of the variables under scrutiny in this examine. It's noticed that every one examined cryptocurrencies exhibit positive returns on common. Notably, Dogecoin is discovered to present the biggest returns but also represents essentially the most volatile digital asset. Cardano and Binance Coin follow in terms of returns whereas Ripple in terms of fluctuations. Emphasis also needs to encompass the actual fact that each one digital currencies current high levels of volatility and kurtosis. High levels of non-regular distribution are also revealed by the Jarque-Bera statistic, presenting evidence that non-linear estimation in quantiles could prove more helpful for identifying causality in relation to standard Granger causality. Moreover, it's noticeable that the DF-GLS and the Phillips-Perron exams point out stationarity of all variables when the primary differences of series are adopted.

Summary statistics.

Note: SD refers to the usual deviation of each asset, JB refers back to the Jarque-Bera statistic, DF-GLS represents the Dickey-Fuller test, while PP refers back to the Phillips-Perron test. ***, ** and * denote important at 1%, 5% and 10% stage respectively.

In Fig. 1, we present the time series evolution of the Twitter-derived economic coverage uncertainty measures and market uncertainty measures throughout the COVID-19 interval. It can be noticed that the most important periods of growth in financial uncertainty appear throughout March 2020, through the section when the number of deaths and admitted patients worldwide was rising exponentially. In the case of market uncertainty measures, they also present very giant phases of development throughout October and November 2020. Further, Fig. 2 illustrates the time collection of the massive-cap cryptocurrencies investigated through the COVID-19 pandemic. It is well discernible that good news concerning the creation of COVID-19 vaccines has provided an incredible enhance available in the market values of cryptocurrencies. This rally started in late 2020 and continued till April 2021.

Twitter-derived financial uncertainty measures and market uncertainty measures during the COVID-19 illness. Note: Within the above Figure, we present the time series evolution of the Twitter-derived economic coverage uncertainty measures and market uncertainty measures during the COVID-19 interval.

Cryptocurrency market values throughout the COVID-19 illness. Note: The above Figure illustrates the time sequence of the massive-cap cryptocurrencies investigated in the course of the COVID-19 pandemic.

The methodology of Diks and Panchenko (2006) is employed for examining whether or not non-linear causality in means exists, deriving from Twitter-based uncertainty indices in the direction of the biggest-cap cryptocurrencies. We first let a strictly bivariate process (X t), Y t, the place X t Granger-causes Y if present values and previous values of the X variable have details about the long run values of the Y variable that are not contained in Y and Y t. Furthermore, we assume that information sets of previous observations of X t and Y t are denoted as F x,t and F y,t. Then X t does not Granger cause Y t when:

Therefore, the null speculation of non-linear Granger causality might be defined as:

Under this null hypothesis, Y t+1 is conditionally unbiased of current and past values of x t, given present and past values of Y t. For finite lags l x and l y, testing the conditional independence could be performed as:

Before applying the Diks and Panchenko (2006) test for non-linear causal effects in means, the speculation of non-linearity should be examined by adopting the BDS take a look at (Broock et al., 1996). If the null hypothesis holds, then the variables examined are identically and independently distributed (i.i.d.) but if the choice hypothesis holds, then there may be linear or non-linear dependency. Moreover, the methodology of Balcilar et al. (2017) is employed for testing non-linear quantile causality in mean and volatility of the cryptocurrencies underneath scrutiny. This methodology is an extension of the work of Nishiyama et al. (2011) and Jeong et al. (2012) and is useful for capturing causality in mean and causality in variance8 . Adopting the non-linear quantile causality specification enables the investigation of the impacts that Twitter-derived financial, or market sentiment exerts on the biggest cryptocurrencies under statement, allowing for the estimation as to whether or not a larger influence is exerted on each digital asset’s mean or volatility. Such a mechanism permits for the thorough investigation of doable paths to enhance portfolio efficiency, significantly as investors are better informed concerning the drivers of danger and returns on such developing belongings. It must be emphasised that the risk-return trade-off could be higher estimated not solely throughout regular circumstances but also when extreme upwards (bull) or downwards (bear) movements develop, as decrease or upper quantile effects might be studied and hidden non-linear results might be more precisely discernible.

The variables y t stand for each of the big-cap cryptocurrencies examined and x t represents the TEU-ENG, TEU-USA, TEU-WGT, TEU-SCA, TMU-ENG, TMU-USA, TMU-WGT, or TMU-SCA Twitter-derived sentiment indicators. Let us suppose that Y t−1 ≡ (y t−1, …, y t−p), X t−1 ≡ (x t−1, …, x t−p), Z t = (X t, Y t) and(F yt∣z t−1)(y t, Z t−1) and(F yt∣y t−1)(y t, Y t−1)are capabilities of the conditional distributions of y t given Z t−1 and Y t−1, respectively.

As Qθ(Zτ−1)≡Qθ(yτ∣Zt−1)andQθ(Yτ−1)≡Qθ(yτ∣Yt−1),thenFyt∣yt−1Q0(Zt−1∣Zt−1)=θˆ with 100% chance. The non-causality hypotheses to be examined are displayed:

The space measure by Jeong et al. (2012), article, more, read here where J = s t E(ε t∣Z t−1)f z(Z t−1) is adopted and εˆt denotes the regression error term whereas f z(Z t−1) denotes the marginal density operate of Z t−1. In Jeong et al. (2012) the possible kernel-based mostly pattern analogue of J follows this form:

It should be noted that K represents the kernel function with bandwidth h, T shows the sample measurement, whereas εtˆ is the estimate of the unknown regression error and is found as:

Notably, Fˆyt∣Yt−a(yy∣Yt−1) constitutes the Nadarya-Watson kernel estimator as beneath:

And the kernel perform is given by L(⋅) whereas the bandwidth is symbolised as h. Additionally, a second-order test is formulated by Balcilar et al. (2017). In line with Nishiyama et al. (2011), the detection of upper-order quantile causality might be achieved by testing:

Thereby, x t Granger causes y t in quantile θ up to the moment ok by using eq. (10) for each ok. Along the traces of Balcilar et al. (2017), testing is conducted regarding non-parametric Granger causality in the first moment (k=1). In case no rejection of the null speculation takes place concerning the primary second nonetheless causality could exist within the second second. That is the explanation why Balcilar et al. (2017) argue that exams in regards to the second moment may nonetheless be utilized. Important conditions for secure quantile causality outcomes are defining the appropriate bandwidth h, the lag order p, as nicely as the Kernel type for K(⋅) and L(⋅). To investigate causality in decrease, medium, and higher elements of the distributions a variety of seven quantiles have been chosen, that's the 0.05, 0.10, 0.25, 0.50, 0.75, 0.90 and 0.95 quantiles9 . This permits us to derive causality in imply and causality in variance in additional detail to effectively estimate the impacts of Twitter-derived financial and market uncertainty on the favored cryptocurrencies under scrutiny.

4. Results

Estimations have been conducted as to whether Twitter-derived financial policy uncertainty measures and market uncertainty measures exert linear Granger results on the biggest cryptocurrencies below statement. Econometric outcomes derived by employing typical Granger causality estimations are displayed in Table 2. It's revealed that conventional Granger causality assessments can not determine causal interlinkages from uncertainty indices in direction of the majority of main cryptocurrencies throughout the COVID-19 pandemic. This phenomenon is even more pronounced in the higher-panel of 2, the place market sentiment impacts are examined. It ought to be underlined that in the examination of Ethereum, Cardano, Ripple, Bitcoin Cash and Litecoin, these property are recognized to be influenced by financial uncertainty as expressed by all tweets within the English language from customers exterior the US. Somewhat surprisingly, there is proof that only Ethereum and Binance Coin are influenced by financial uncertainty in the US. The weighted index and the scaled index led to related findings as only Ethereum, Binance Coin and Bitcoin Cash are receivers of causal effects in both case. In the decrease panel of Table 2, linear Granger causality is sort of non-existent, as Binance Coin is the only main cryptocurrency receiving Granger results by Twitter market uncertainty exterior of the US. The remaining market uncertainty measures are revealed not to be sources of causal impacts in any examined cryptocurrency.

Linear Granger Causality Tests.

Note: The null hypothesis of non-linear Granger causality might be outlined as: H0:Xt shouldn't be Granger causing Yt; where beneath this said null speculation, Yt+1 is conditionally independent of current and previous values of xt, given present and past values of Yt. ***, ** and * denote significant at 1%, 5% and 10% stage respectively.

to analyze as to whether or not the non-linear method is an applicable mechanism to detect causality from uncertainty measures to digital currencies, we apply the BDS test (Broock et al., 1996) on the residuals of the returns equation within the Vector Autoregressive (VAR) framework. Results are presented in Table 3 and supply clear proof that in every variable examined, the null speculation of no serial dependence across a range of dimensions is rejected. This signifies that non-linearity exists in a statistically vital manner even at the 99% confidence level.

BDS results.

Note: The appropriateness of the non-linear method is examined utilizing the BDS check (Broock et al., 1996) on the residuals of the returns equation in the Vector Autoregressive (VAR) framework. ***, ** and * denote significant at 1%, 5% and 10% degree respectively.

The methodology of Diks and Panchenko (2006) is subsequent employed to analyze non-linear causality in means of cryptocurrencies. The econometric outcomes derived are specified by Table 4. Presented evidence offers credence to the notion that only some of the largest-cap digital currencies are receivers of non-linear Granger causality in imply by Twitter-derived uncertainty measures. To be extra precise, it is revealed that Bitcoin, Ethereum, Bitcoin Cash and Litecoin are non-linearly Granger-caused in mean by all 4 Twitter-derived financial coverage uncertainty measures. Moreover, Stellar is influenced solely by the non-US index (TEU-ENG) and the scaled index (TEU-SCA). Furthermore, Cardano and Ripple are found to be receivers of Granger causality in mean only at the second dimension (m = 2). These findings reveal that nearly all of an important and properly-established digital currencies (Bitcoin, Ethereum, Bitcoin Cash, and Litecoin) react to shocks in Twitter-derived financial coverage uncertainty. This documents the existence of tighter linkages between extra environment friendly cryptocurrency markets and fashionable sentiment indicators in comparison with less established, massive cryptocurrency markets. This coincides with the view that as modern monetary tools, they share extra frequent characteristics with traditional belongings, and represent a part of the overall economic atmosphere to a larger extent.

Nonlinear Granger causality check, Diks and Panchenko (2006).

Note: The Diks and Panchenko (2006) methodology is employed for the purpose of investigating non-linear causality within the imply values of our chosen cryptocurrencies. ***, ** and * denote vital at 1%, 5% and 10% degree respectively. m demonstrates the embedding dimension.

Alternatively, it should be underlined that proof indicates only Ethereum and Bitcoin Cash are receivers of non-linear results in imply, no matter which Twitter-based mostly market uncertainty takes place. Outcomes current that the non-US, the US, and the weighted market uncertainty sentiment exert impact upon Cardano in sure dimensions. Additionally, the weighted and the scaled indices are sources of affect upon Stellar (however not in all dimensions examined), and the only weighted index exerts causal impacts on the mean of the Litecoin cryptocurrency (in two dimensions). These outcomes strengthen the view that only the closest substitutes of Bitcoin could show wealth-generating and be intensely susceptible to market investor sentiment10 .

Aside from econometric procedures to trace the non-linear causality within the imply using the methodology of Diks and Panchenko (2006), extra advanced estimations are performed to examine whether or not non-linear causality in mean in addition to in volatility exists. This task has been undertaken by adopting the Balcilar et al. (2017) econometric specification11 Furthermore and for the sake of completeness, we doc Galvao’s quantile unit root take a look at at stage (Galvao, 2009). The findings are tabulated in Table 5 which suggests that null hypothesis of unit root cannot be rejected on the 5% significance stage for almost all the variables and quantiles. Although, quantile unit root test verifies stationarity for variables in first variations and that is according to the estimations12 of the unit root checks in Table 1. The outcomes of Balcilar et al. (2017) econometric specification are illustrated in Figs. 3, 4, 5, 6, 7, 8, 9 by 10. More precisely, in Figures the important values (CV) at 10%, 5% and 1% stage of significance is 1.645, 1.96, and 2.33 respectively. The vertical axis of the graphs depicts the worth of quantile causality in mean and in variance (test statistic) while the horizontal axis represents the level of quantile distribution (q = 0.05, 0.10, 0.15, 0.20, 0.25, …, 0.80, 0.85, 0.90, 0.95). Findings present evidence that the Twitter-derived economic and market sentiment indices exert negligible impacts on the means of every of the cryptocurrencies under scrutiny. Notably, this is available in stark contrast with outcomes primarily based on non-linear Granger impacts on volatilities. It is clear from such results, that all of the eight uncertainty measures affect the volatilities of Bitcoin, Ethereum, Binance Coin, Dogecoin, Bitcoin Cash, Litecoin, and Ethereum Classic in a non-linear method in any respect quantiles investigated. Nevertheless, it ought to be famous that volatility within the Cardano market is just not affected at the bottom quantile, while Ripple’s volatility does not receive impacts at the 2 lowest and the upper quantile. Non-linear volatility impacts being current only in center quantiles is more evident within the case of Stellar were the three lowest, and the upper quantile are revealed to be unaffected by Twitter-primarily based uncertainty during the COVID-19 period. It's value mentioning that the nice majority of statistically significant estimations take pleasure in excessive levels of statistical reliability (they're valid on the 99% confidence stage). When it comes to estimations of non-linear causality at the higher quantile although most estimations are hardly important at the 95% confidence degree.

Quantile Unit Root check.

Note: In the above Table, we doc Galvao’s quantile unit root test at level (Galvao, 2009). * shows rejection of the null speculation on the 5% significance stage.

Quantile causality-in-mean and in-variance for TEU-ENG. Note: The important values (CV) at 1%, 5% and 10% degree of significance is 2.33. 1.96, and 1.645 respectively.

Quantile causality-in-imply and in-variance for TEU-USA. Note: The essential values (CV) at 1%, 5% and 10% level of significance is 2.33. 1.96, and 1.645 respectively.

Quantile causality-in-imply and in-variance for TEU-WGT. Note: The critical values (CV) at 1%, 5% and 10% stage of significance is 2.33. 1.96, and 1.645 respectively.

Fig. 6.

Quantile causality-in-imply and in-variance for TEU-SCA. Note: The critical values (CV) at 1%, 5% and 10% level of significance is 2.33. 1.96, and 1.645 respectively.

Fig. 7.

Quantile causality-in-imply and in-variance for TMU-ENG. Note: The essential values (CV) at 1%, 5% and 10% level of significance is 2.33. 1.96, and 1.645 respectively.

Fig. 8.

Quantile causality-in-mean and in-variance for TMU-USA. Note: The crucial values (CV) at 1%, 5% and 10% degree of significance is 2.33. 1.96, and 1.645 respectively.

Fig. 9.

Quantile causality-in-imply and in-variance for TMU-WGT. Note: The important values (CV) at 1%, 5% and 10% degree of significance is 2.33. 1.96, and 1.645 respectively.

Fig. 10.

Quantile causality-in-mean and in-variance for TMU-SCA. Note: The vital values (CV) at 1%, 5% and 10% degree of significance is 2.33. 1.96, and 1.645 respectively.

Emphasis should be given on that the three cryptocurrencies not affected by Twitter-derived uncertainty, both emanating from economic circumstances or market situations, constitute digital belongings with low market values. Arguably, Cardano, Ripple, and Stellar are cryptocurrencies with excessive levels of volatility and this has allowed buyers to speculate by holding and promoting them despite the fact that their nominal worth stays low. A standard characteristic of every is the presence of price fluctuation of costs, even when cryptocurrency markets remain somewhat inactive when it comes to buying and selling activity. These cryptocurrencies show low sensitivity to economic uncertainty in addition to to market uncertainty. That is the explanation why they're unaffected by Twitter-related uncertainty measures and stay partially intact as the COVID-19 pandemic develops and persists. These outcomes present course for traders, indicating that low-nominally valued cryptocurrencies should not vulnerable to extreme movements, through the financial crisis to a bigger extent than in normal instances.

5. Conclusions

This research investigates non-linear causality in both the mean and volatilities of the biggest cryptocurrencies as sourced from Twitter-derived economic and market uncertainty throughout the COVID-19 pandemic. Econometric outcomes primarily based upon non-linear (quantile) methodologies are employed to offer a better understanding of the determinants of this novel, popular speculative investments are offered in each normal and recognized bull and bear market conditions.

Results counsel that Bitcoin, Ethereum, Bitcoin Cash, and Litecoin are non-linearly influenced of their respective imply by the selected Twitter-derived economic uncertainty indices in a statistically important manner. Then again, only Ethereum, Bitcoin Cash, and partially Cardano are found to be non-linearly attributable to Twitter-primarily based market uncertainty measures. When results of non-linear quantile causality are analysed, it is well discernible that Bitcoin, Ethereum, Binance Coin, Dogecoin, Bitcoin Cash, Litecoin, and Ethereum Classic are receivers of impacts at all quantiles in a statistically dependable manner. It must be underlined though that almost all digital currencies with low nominal market values (specifically Cardano, Ripple, and Stellar) remain unaffected by Twitter-derived sentiment indicators at the lowest or highest quantiles of their volatilities. This consequence may be partially defined as such low-priced cryptocurrencies very often present modest levels of volatility, even in occasions when modest ranges of economic uncertainty or investor optimism exist. Such digital currencies are found to be unaffected by unstable investor sentiment or by monetary crises.

Such a phenomenon could be very useful for hedging functions in portfolios that include fashionable or traditional financial property. Such results find that Twitter-derived sentiment measures are unable to clarify the recognized volatility in low nominally-priced cryptocurrencies, even in intensely distressed periods such as the COVID-19 pandemic, thought of to be a significant worldwide financial disaster. Thereby, these cryptocurrencies can function protected havens throughout turbulent financial and market situations. The alternative methodology indicates that non-linear causality-in-imply exist in digital currencies that present larger resemblances with traditional assets behavioural traits. Therefore, it can be concluded that traders prepared to hedge their portfolios from the effects of the pandemic would profit by investing in low nominally-priced, yet extremely-capitalised cryptocurrencies. Potential avenues for additional research encompass additional investigation of superior types of causal results exercised by even more refined investor sentiment measures.

Resembling: "economic, economical, economically, economics, economies, economist, economists, economy" and uncertainty similar to: "uncertainty, unsure, uncertainties, uncertainly"

Such work built on that developed by Corbet et al. (2021) and Umar et al. (2021)

Other relevant pieces embrace Diniz-Maganini et al. (2021), Goodell and Goutte (2021), Guo et al. (2021), Huang et al. (2021), Corbet et al. (2021) and Shehzad et al. (2021).

It is argued that the nexus of Bitcoin with traditional belongings has altered through the pandemic as well as with particular segments whereas the reference to the US makes the exception (Corbet et al., 2020, Corbet et al., 2020, Corbet et al., 2021). Shehzad et al. (2021) utilise Morlet Wavelet method and present that gold is preferable to Bitcoin during the pandemic relating to its secure haven talents (Corbet et al., 2020). It's supported that the Gold/Bitcoin ratio elevated in the majority of the Asian, European and US markets investigated.

Available from the www.policyuncertainty.comwebpage

Furthermore, ENG signifies all tweets in the English language from customers exterior the US,while USA stands for all tweets made in the United States. Additionally, WGT is the weighted index whereas SCA illustrates the scaled index.

Data has been obtained from the Coinmarketcap database

That is certainly one of the important thing explanation why tracing causality between variables is carried out extra precisely in comparison with standard Granger causality methodologies.

Methodological variants, equivalent to constructing on the work of Galvao (2009), and using lower denominations of quantiles had been thought-about, nevertheless, for brevity solely these stated have been presented. The outcomes of substantial further analysis can be found from the authors upon request.

A lot of traders consider that the mature Bitcoin market may have reached its peak and will not have the ability to show great returns sooner or later. That is the rationale why elevated investor attention has centered on the closest substitutes of Bitcoin that could provide credibility but additionally stay eligible for brand new bubble creations over time.

It must be underlined that the outcomes extracted about the impacts of TEU-ENG, TEU-USA, TEU-WGT, TEU-SCA, TMU-ENG, TMU-USA, TMU-WGT, and TMU-SCA on the ten largest-cap cryptocurrencies bear massive similarities

Detailed results are omitted for brevity of presentation, and can be found from the authors upon request.

1. Akyildirim E., Cepni O., Corbet S., Uddin G. Forecasting mid-price motion of Bitcoin futures utilizing machine studying. Annals of Operations Research. 2021 doi: 10.1007/s10479-021-04205-x. [DOI] [PMC free article] [PubMed] [Google Scholar]2. Akyildirim E., Corbet S., Cumming D., Lucey B., Sensoy A. Riding the wave of crypto-exuberance: The potential misusage of corporate blockchain bulletins. Technological Forecasting and Social Change. 2020;159 [Google Scholar]3. Akyildirim E., Corbet S., Lucey B., Sensoy A., Yarovaya L. The connection between implied volatility and cryptocurrency returns. Finance Research Letters. 2020;33 [Google Scholar]4. Akyildirim E., Corbet S., Sensoy A., Yarovaya L. The affect of blockchain associated name changes on corporate efficiency. Journal of Corporate Finance. 2020;Sixty five [Google Scholar]5. Akyildirim E., Sensoy A., Gulay G., Corbet S., Salari H. Big data analytics, order imbalance and the predictability of inventory returns. Journal of Multinational Financial Management. 2021;Sixty two [Google Scholar]Baker, S. R., Bloom, N., Davis, S., Renault, T. (2021). Twitter-derived measures of financial uncertainty.6. Balcilar M., Bekiros S., Gupta R. The position of reports-based mostly uncertainty indices in predicting oil markets: A hybrid non-parametric quantile causality technique. Empirical Economics. 2017;53(3):879-889. [Google Scholar]7. Béjaoui A., Mgadmi N., Moussa W., Sadraoui T. A brief-and lengthy-time period analysis of the nexus between Bitcoin, social media and COVID-19 outbreak. Heliyon. 2021 doi: 10.1016/j.heliyon.2021.e07539. [DOI] [PMC free article] [PubMed] [Google Scholar]8. Beneki C., Koulis A., Kyriazis N.A., Papadamou S. Investigating volatility transmission and hedging properties between Bitcoin and Ethereum. Research in International Business and Finance. 2019;48:219-227. [Google Scholar]9. Broock W.A., Scheinkman J.A., Dechert W.D., LeBaron B. A take a look at for independence based on the correlation dimension. Econometric Reviews. 1996;15(3):197-235. [Google Scholar]10. Cioroianu I., Corbet S., Larkin C. The differential impact of company blockchain-development as conditioned by sentiment and financial desperation. Journal of Corporate Finance. 2021:66. [Google Scholar]11. Conlon T., Corbet S., McGee R.J. Are cryptocurrencies a safe haven for fairness markets? An international perspective from the COVID-19 pandemic. Research in International Business and Finance. 2020;Fifty four doi: 10.1016/j.ribaf.2020.101248. [DOI] [PMC free article] [PubMed] [Google Scholar]12. Corbet S., Hou G., Hu Y., Oxley L. We Reddit in a discussion board: The affect of messaging boards on firm stability. Review of Corporate Finance. 2021 [Google Scholar]13. Corbet S., Hou Y., Hu Y., Larkin C., Lucey B., Oxley L. Cryptocurrency liquidity and volatility interrelationships throughout the COVID-19 pandemic. Finance Research Letters. 2021 doi: 10.1016/j.frl.2021.102137. [DOI] [PMC free article] [PubMed] [Google Scholar]14. Corbet S., Hou Y., Hu Y., Larkin C., Oxley L. Any port in a storm: Cryptocurrency protected-havens throughout the COVID-19 pandemic. Economics Letters. 2020:194. doi: 10.1016/j.econlet.2020.109377. [DOI] [PMC free article] [PubMed] [Google Scholar]15. Corbet S., Hou Y., Hu Y., Lucey B., Oxley L. Aye Corona! The contagion results of being named Corona through the COVID-19 pandemic. Finance Research Letters. 2021:38. doi: 10.1016/j.frl.2020.101591. [DOI] [PMC free article] [PubMed] [Google Scholar]16. Corbet S., Hou Y., Hu Y., Oxley L. The influence of the COVID-19 pandemic on asset-price discovery: Testing the case of Chinese informational asymmetry. International Review of Financial Analysis. 2020:72. doi: 10.1016/j.irfa.2020.101560. [DOI] [PMC free article] [PubMed] [Google Scholar]17. Corbet S., Larkin C., Lucey B. The contagion effects of the COVID-19 pandemic: Evidence from gold and cryptocurrencies. Finance Research Letters. 2020:35. doi: 10.1016/j.frl.2020.101554. [DOI] [PMC free article] [PubMed] [Google Scholar]18. Diks C., Panchenko V. A new statistic and sensible guidelines for non-parametric Granger causality testing. Journal of Economic Dynamics and Control. 2006;30(9-10):1647-1669. [Google Scholar]19. Diniz-Maganini N., Diniz E.H., Rasheed A.A. Bitcoinas worth effectivity and protected haven properties through the COVID-19 pandemic: A comparability. Research in International Business and Finance. 2021 doi: 10.1016/j.ribaf.2021.101472. [DOI] [PMC free article] [PubMed] [Google Scholar]20. Dyhrberg A., Foley S., Svec J. How investible is Bitcoin? Analyzing the liquidity and transaction prices of Bitcoin markets. Economics Letters. 2018;171:140-143. [Google Scholar]21. Eross A., McGroarty F., Urquhart A., Wolfe S. The intraday dynamics of Bitcoin. Research in International Business and Finance. 2019;49:71-81. [Google Scholar]22. Fang T., Su Z., Yin L. Economic fundamentals or investor perceptions? The role of uncertainty in predicting long-term cryptocurrency volatility. International Review of Financial Analysis. 2020;71 [Google Scholar]23. Fletcher E., Larkin C., Corbet S. Countering money laundering and terrorist financing: A case for bitcoin regulation. Research in International Business and Finance. 2021:56. [Google Scholar]24. Fry J. Booms, busts and heavy-tails: The story of Bitcoin and cryptocurrency markets? Economics Letters. 2018;171:225-229. [Google Scholar]25. Galvao A.F., Jr Unit root quantile autoregression testing using covariates. Journal of Econometrics. 2009;152(2):165-178. [Google Scholar]26. Gandal N., Hamrick J., Moore T., Vasek M. The rise and fall of cryptocurrency coins and tokens. Decisions in Economics and Finance. 2021;44(2):981-1014. [Google Scholar]27. Goodell J.W., Goutte S. Co-movement of COVID-19 and Bitcoin: Evidence from wavelet coherence analysis. Finance Research Letters. 2021;38 doi: 10.1016/j.frl.2020.101625. [DOI] [PMC free article] [PubMed] [Google Scholar]28. Gozgor G., Tiwari A.K., Demir E., Akron S. The relationship between Bitcoin returns and commerce coverage uncertainty. Finance Research Letters. 2019;29:75-82. [Google Scholar]29. Guo X., Lu F., Wei Y. Capture the contagion network of bitcoin-Evidence from pre and mid COVID-19. Research in International Business and Finance. 2021;58 doi: 10.1016/j.ribaf.2021.101484. [DOI] [PMC free article] [PubMed] [Google Scholar]30. Huang Y., Duan K., Mishra T. Is Bitcoin really greater than a diversifier? A pre-and submit-COVID-19 evaluation. Finance Research Letters. 2021 [Google Scholar]31. Huynh T.L.D. Does Bitcoin react to Trumpas Tweets? Journal of Behavioral and Experimental Finance. 2021;31 doi: 10.1016/j.jbef.2021.100536. [DOI] [PMC free article] [PubMed] [Google Scholar]32. Jeong K., Härdle W.K., Song S. A constant non-parametric test for causality in quantile. Econometric Theory. 2012;28(4):861-887. [Google Scholar]33. Katsiampa P. Volatility estimation for Bitcoin: A comparison of GARCH fashions. Economics Letters. 2017;158:3-6. [Google Scholar]34. Katsiampa P., Corbet S., Lucey B. High frequency volatility co-movements in cryptocurrency markets. Journal of International Financial Markets, Institutions and Money. 2019;62:35-52. [Google Scholar]35. Katsiampa P., Corbet S., Lucey B. Volatility spillover effects in leading cryptocurrencies: A BEKK-MGARCH analysis. Finance Research Letters. 2019;29:68-74. [Google Scholar]36. Kraaijeveld O., De Smedt J. The predictive power of public Twitter sentiment for forecasting cryptocurrency prices. Journal of International Financial Markets, Institutions and Money. 2020;Sixty five [Google Scholar]37. Kyriazis N., Papadamou S., Corbet S. A systematic evaluate of the bubble dynamics of cryptocurrency prices. Research in International Business and Finance. 2020;Fifty four [Google Scholar]38. Lee A.D., Li M., Zheng H. Bitcoin: Speculative asset or innovative expertise? Journal of International Financial Markets, Institutions and Money. 2020;67 [Google Scholar]39. Lehrer S., Xie T., Zhang X. Social media sentiment, mannequin uncertainty, and volatility forecasting. Economic Modelling. 2021 [Google Scholar]40. Li Y., Goodell J.W., Shen D. Comparing search-engine and social-media attentions in finance research: Evidence from cryptocurrencies. International Review of Economics & Finance. 2021;75:723-746. [Google Scholar]41. Naeem M.A., Mbarki I., Shahzad S.J.H. Predictive role of on-line investor sentiment for cryptocurrency market: Evidence from happiness and fears. International Review of Economics & Finance. 2021;73:496-514. [Google Scholar]42. Nishiyama Y., Hitomi K., Kawasaki Y., Jeong K. A consistent nonparametric test for nonlinear causality-Specification in time collection regression. Journal of Econometrics. 2011;165(1):112-127. [Google Scholar]43. Papadamou S., Kyriazis N., Tzeremes P., Corbet S. Herding behaviour and value convergence clubs in cryptocurrencies during bull and bear markets. Journal of Behavioral and Experimental Finance. 2021:30. [Google Scholar]44. Philippas D., Rjiba H., Guesmi K., Goutte S. Media attention and Bitcoin costs. Finance Research Letters. 2019;30:37-43. [Google Scholar]45. Raheem I.D. COVID-19 pandemic and the secure haven property of Bitcoin. The Quarterly Review of Economics and Finance. 2021;81:370-375. [Google Scholar]46. Sarkodie S.A., Ahmed M.Y., Owusu P.A. COVID-19 pandemic improves market indicators of cryptocurrencies-proof from Bitcoin, Bitcoin Cash, Ethereum, and Litecoin. Finance Research Letters. 2021 doi: 10.1016/j.frl.2021.102049. [DOI] [PMC free article] [PubMed] [Google Scholar]47. Sensoy A., Silva T., Corbet S., Tabak B. High-frequency return and volatility spillovers amongst cryptocurrencies. Applied Economics. 2021;53(37):4310-4328. [Google Scholar]48. Shehzad K., Bilgili F., Zaman U., Kocak E., Kuskaya S. Is gold favourable to bitcoin throughout the COVID-19 outbreak? Comparative analysis via wavelet approach. Resources Policy. 2021 doi: 10.1016/j.resourpol.2021.102163. [DOI] [PMC free article] [PubMed] [Google Scholar]49. Tumasjan A., Braun R., Stolz B. Twitter sentiment as a weak sign in venture capital financing. Journal of Business Venturing. 2021;36(2) [Google Scholar]50. Umar Z., Gubareva M., Yousaf I., Ali S. A tale of firm fundamentals vs sentiment pushed pricing: The case of GameStop. Journal of Behavioral and Experimental Finance. 2021;30 [Google Scholar]51. Wang P., Li X., Shen D., Zhang W. How does financial coverage uncertainty affect the Bitcoin market? Research in International Business and Finance. 2020;Fifty three [Google Scholar]52. Wu W., Tiwari A.K., Gozgor G., Leping H. Does financial coverage uncertainty have an effect on cryptocurrency markets? Evidence from Twitter-primarily based uncertainty measures. Research in International Business and Finance.