“Analysis of the impact of central bank digital currency on stock markets: Dynamics and implications”

The purpose of the study is to explore the influence of central bank digital currency on stock markets. To realize the purpose, the TVP-VAR model was built, which determines the impact of volatility of the CBDC attention index (CBDCAI) on the volatility of stock market indices. The study uses a time-varying vector autoregressive model that analyzes weekly data from the first week of January 2015 to the first week of July 2023. The endogenous vector to be assessed by VAR contains CBDCAI and stock market indices of different countries (France: CAC 40, The United States of America: S&P 500, Germany: DAX 40, United Kingdom: FTSE 100, China: SSEC, The Netherlands AEX 25, Switzerland: SMI 20, Japan: Nikkei 225, India: NIFTY 50, Brazil: BVSP, South Korea: KOSPI). The results of the TVP-VAR model show that compared to stock market indices, CBDCAI appeared to be relatively independent and isolated. Interdependence and mutual influence between the digital currency market of central banks and stock markets were also revealed. In addition, CBDC functions primarily as a volatility absorber rather than a source of volatility. Despite the overall ability of the CBDC market to absorb fluctuations in volatility, it may also change its function with the widespread adoption of central bank digital currencies in many countries.


INTRODUCTION
The modern financial world is experiencing a period of stunning innovation and change.One of the important financial innovations is the introduction of central bank digital currencies (CBDCs).CBDC is created digitally and recorded in centralized or decentralized ledgers.It is expressed in the existing unit of account and can be exchanged for physical cash, money from commercial banks and other forms of money at the owner's request through authorized institutions.If CBDC is introduced, it will become legal tender, just like physical cash.
In recent years, central banks of various countries have been actively engaged in research and development in the field of central bank digital currency and are competing for leadership in this area.The launching of the world's first CBDC, known as "Sand Dollar", took place in the Bahamas in 2020, and this event officially marked a new era in the history of money.After its launch, the digital currency became very popular and generated a lot of interest (Skrynnyk & Lyeonov, 2022;Liu, 2023;Bublyk et al., 2023;Dubyna et al., 2023).As of September 2023, a significant increase in interest in the implementation of a central bank digital currency was recorded in the world.Currently, 130 countries, which together represent 98% of global GDP, are actively exploring the possibility of implementing CBDC (Atlantic Council, 2023).
There is currently considerable uncertainty in the financial industry about how exactly CBDCs affect stock markets and their volatility.Wang et al. (2022) studied the impact of the central bank digital currency attention index (CBDCAI) on financial markets.The results of this study showed a possible relationship between the volatility of the CBDAI and the volatility of the FTSE All World index.
The main innovation of the study is the use of quantitative analysis to study the impact of volatility of the CBDC attention index on the stock market volatility.A key question is whether CBDCs have the potential to change the markets' perception of news, affect the level of volatility and change the performance of stock markets.The next important question is to determine whether the CBDCAI is a source of stock market volatility or whether the index absorbs volatility.Therefore, the study aims to prove the hypothesis that there is a significant relationship between the volatility of the CBDCAI and the volatility of the stock market.
The problem requires a detailed analysis of the interaction between central bank digital currencies and stock markets to understand their impact on global financial stability and the ability of investors to make informed decisions.

LITERATURE REVIEW
The trends in recent years indicate an increase in interest in digital currencies, especially due to the attempt by private organizations to reproduce the characteristics of cash in a digital environment.Digital currencies exist in different formats and value propositions and can be used on a variety of platforms for real-time transactions.
Investigating this new phenomenon, the Committee on Payments and Market Infrastructures determined that digital currencies have certain characteristics of a currency, commodity, or asset.The use of digital currencies for retail payment services may be particularly important, as they can facilitate e-commerce, cross-border transactions, and P2P payments, making them faster and more accessible (World Bank Group, 2021).
As of September 2023, 64 countries are in the advanced research phase, meaning they are in the process of developing, piloting or even launching their CBDCs.In particular, 19 countries from the G20 are actively working on the development of CBDC, and among them 9 countries already have pilot projects in this area.Over the past six months, each of these G20 countries has taken a significant step forward and allocated new resources to implement these projects (Atlantic Council, 2023).
Fully launched digital currencies already exist in 11 countries.China, for example, has a CBDC pi-lot project that currently includes 260 million users and is being tested in more than 200 scenarios, including the use in public transportation, incentive payments and e-commerce (Atlantic Council, 2023).
Countries will implement different forms of central bank digital currency systems depending on their specific goals.These systems can be divided into two main categories: single ledger systems and distributed ledger systems.In systems with single ledger systems, the central bank and other government agencies are responsible for CBDC management.This model may have a single management system, but it also considers the difference between a central bank and a management account.In systems based on a distributed ledger, management is carried out jointly by several parties, which promotes transparency of operations and reduces the risk of digital attacks.Such systems use distributed accounts that do not disclose personal information to verify the identity of users (Liu, 2023).One important feature is that CBDC, as a central bank liability, can be considered riskfree and irrevocable (Boar & Wehrli, 2021).
CBDC systems can also be classified by the type of central bank control as direct or indirect.In a direct CBDC system, the central bank directly controls the issuance and management of CBDC.In the case of a non-direct CBDC system, partial authority may be delegated to intermediary agencies, but the management of intermediaries remains with the central bank.These different approaches to the implementation of CBDC have their advantages and disadvantages and correspond to different goals and contexts that countries choose according to their needs and circumstances (Chu et al., 2022).
The implementation of CBDC will have a serious impact on countries' financial systems and will require careful monitoring and analysis.The future financial stability will require deep interaction between the financial system and digital technology systems.The stability of the financial system can be achieved by jointly ensuring the stability of financial institutions and the reliable functioning of the digital technology system on which financial institutions rely (Ozili, 2022b;Kerimov et al., 2023;Panigrahi, 2023).
Research has also indicated that the introduction of a central bank digital currency has the potential to affect financial markets and systems in different ways.The general support for sustainable development, public safety, and privacy (as a positive factor) will contribute to the positive perception of the digital currency of the central bank by society.This is important because one of the key risk situations that central banks can face is the possibility of launching (retail) CBDC projects that fail to gain popularity among mass users (Passacantando, 2021).
To avoid this risk, central banks should include in their strategy all aspects reviewed from the position of a broad understanding of sustainability.This includes minimizing energy consumption, ensuring social integration through the availability of CBDC to a wide range of users, significantly reducing the cost of money transfers and fighting financial crime.
Central banks' digital currency privacy will be tested through litigation.To facilitate this process, each CBDC project should be underpinned by a specific legal foundation so that courts can appropriately evaluate the invasion of privacy.With these factors in mind, central banks must adapt their strategy to develop CBDC projects that meet the requirements of sustainability, public security, and privacy.This approach will help reduce the environmental impact of CBDCs and gain social recognition for these projects (Rybski, 2023) This means that wholesale CBDCs can facilitate faster and more efficient inter-bank transactions, reducing reliance on traditional payment systems.On the other hand, retail CBDCs that are available to the general public could have broader implications.Also, CBDC issuance can have an important impact on central bank balance sheets depending on the modality of CBDC conversion (Kiff et al., 2020).This means that central banks may need to adjust their balance sheets and operations to accommodate the implementation of CBDC.Additionally, CBDCs can potentially increase financial accessibility by expanding banks' depositor base if additional competition forces banks to raise deposit rates (Ozili, 2022a).This can lead to a more inclusive banking system as more people will have access to banking services.CBDCs could also change the role of central banks in the financial system.They can give central banks more direct control over monetary policy and financial stability (Fung & Halaburda, 2016).With the help of CBDC, central banks can see transactions in real time and potentially implement more targeted and effective monetary policy.CBDC can also improve the efficiency and security of payments, reducing reliance on intermediaries and increasing the resilience of the financial system (Kshetri, 2021).However, the introduction of CBDCs can also present challenges and risks.For example, the impact of CBDC on private banks needs to be carefully assessed.An interest-bearing CBDC could potentially affect the profitability and busi-ness model of private banks, especially in the monopoly banking sector (Andolfatto, 2018).Central banks need to consider the potential consequences and develop appropriate policies to mitigate any negative effects on private banks.
CBDCs have the potential to transform the banking system and central banks.They can improve the efficiency and inclusiveness of the financial system, increase the effectiveness of monetary policy, and strengthen the stability of payments.However, implementing a CBDC also requires careful consideration of the impact on private banks and the necessary adjustments to central bank operations and balance sheets.

DATA DESCRIPTION AND METHODOLOGY
Weekly data from January 2015 to September 2023 was collected to analyze the impact of central bank digital currency attention index volatility on stock markets.This starting point in the analysis was chosen due to the availability of CBDCAI attention index, which was calculated (Wang et at, 2022) and is available on a weekly basis.We believe that using weekly data will increase the number of observations.In addition, the use of TVP-VAR model with a variable structure of the variance-covariance matrix will make it possible to consider the volatile nature of high-frequency financial variables, as indicated by Primiceri et al. (2005).
The next stage is to decide on the inclusion of stock market indices in the study, taking into account the main task of the study, namely the analysis of the impact of volatility of the central bank's digital currency attention index on the volatility of stock market indices.The selection of equity markets for inclusion in the analysis is based on the previous research that has shown which markets responded to CBDC-related disruptions and which remained resilient to these disruptions (Kuznichenko, 2022) For the study, an empirical analysis will be conducted to examine the impact of CBDCAI volatility on international stock market indices (Appendix A, Table A1).
Stock exchange indices of countries where the state of CBDC launching is in different phases (research, pilot launching, launching) were selected for the study.
For the purpose of this analysis, the Therefore, when evaluating the TVP-VAR model, the following vector of endogenous variables is considered (Figure 1):   where u t is covariance matrix ∑ u of a k-dimensional unobservable process with zero mean vector white noise; ϕ i = JA i J i and J = [I k : 0: 0: ... 0]; A i -can be summed up.
Using TVP-VMA, IRF can identify a shock to a single variable and trace its marginal effect.Therefore, according to Wang (2022), IRF for each variable j on variable i can be calculated as follows: ( ) where e i ՛ and e j are two fundamental (N × 1) dimensional vectors, each with a unit i and j, respectively.(A = K × K p ) is a matrix of coefficients for time-varying parameters.In time period (tp), the response impulse equals the cumulative forecast error from the shock for variables i and j.
The next step of the study is to predict the error variance for the k-th element of the forecast error vector.For this purpose, Wang (2022) introduced the following formula: ( where θ jk,0 2 + ... + θ jk,h-1 2j th ε t is the contribution of innovation to the variance of the h-step forecast error of variable k. ω kj,h -j th ε t contribution of innovation to the error variance of the h-step k. After clarifying the mathematical basis of volatility spread, the Total Spillover Connectivity Index (TSCI) can now be derived from the above equation based on FEVD (Wang, 2022).
The analysis of TSCI allows us to consider the dynamics of the relationship between different variables in the system.The paper examines the relationship between CBDCAI and stock markets, which were described earlier.This approach is similar to the one used in the analysis of systemic shocks.For example, unit "A" has the largest momentum and can transmit this momentum to nearby units.Then the impulses spread to neighboring units and so on.In the propagation process, high values can spread quickly, while low values can spread slowly and weaken (Kuznichenko, 2022).
The equations for calculating FEVD and TSCI provide an opportunity to determine the directional connectivity spillover (DCS).Directional spillover occurs when variable "i" receives information from all other variables in the system and simultaneously transmits information to all other variables in the system.DCS can be described as TSCI decomposition process "from" or "to" a particular data source.There are four different ways to measure DCS: 1) DCS f -connectivity from spillover; 2) DCS t -connectivity to spillover; 3) DCS n -connectivity to network spillover; 4) DCS np -connectivity of pairwise directed spillover.
It should be noted that DCS n between variables i and j measures the difference between the spillovers transmitted from i and j and the spillovers from j and i.A moving panel estimation regression methodology was applied to generate time-dependent schedules of these four DSC indicators to capture the dynamic relationships between all variables.Accordingly, the following formulas were used (Wang, 2022): ,

Descriptive analysis
The provided statistical data (Table 1) give an idea of the characteristics of various financial indicators.Each column represents a specific statistical indicator, and their combined analysis can help draw broader conclusions about the data set.
The mean values in the data set vary significantly, indicating different mean levels for each financial indicator.For example, the mean value for BVSP is significantly higher than other indices, suggesting a different scale for the Brazilian stock market index.
Standard deviations (Std) provide insight into the variance or variability of data.Indices such as FCHI and BVSP have relatively high standard deviations, indicating greater variability in these indices.
The range (the difference between the maximum and minimum values) is significant for some indices such as BVSP and N225.This indicates that these financial indicators have a wide range of values during the observation period.
Skewness measures the asymmetry of data distribution.Indices with positive skewness have a longer tail on the right side of the distribution, while columns with negative skewness have a longer tail on the left side.Notably, CBDCAI and SSEC exhibit positive skewness, while FCHI and BVSP exhibit negative skewness.
Kurtosis measures the granularity of distribution.
Positive kurtosis indicates heavier tails, while negative kurtosis indicates lighter tails.SSEC has a significantly positive kurtosis, implying fat tails in its distribution.
The IQR is a measure of the spread of data between the 25th and 75th percentiles.Indices with a larger IQR have a larger distribution of data between the lower and upper quartiles.FCHI and BVSP have significantly higher IQRs, indicating a wider range of values in the middle 50% of the data.
The Jarque-Bera test assesses whether the data follows a normal distribution.A high Jarque-Bera statistic indicates a significant deviation from the norm.Indices with larger Jarque-Bera statistics, such as CBDCAI, SSEC, and BVSP, deviate more from a normal distribution.
It should be noted that all columns have the same number of records (count -444), which indicates stable data availability.
Some financial indicators show significant differences in scale, variability, and form of distribution.For example, the BVSP stands out with a much higher mean, standard deviation and range compared to other stock market indices.
The next step analyzed the results (Table 2) of both the Augmented Dickey-Fuller (ADF) and Phillips-Perron stationarity tests for each index and CBDCAI.
The results show that the ADF test statistic is highly negative for all indices, indicating a strong rejection of the null hypothesis of non-stationarity.This is further supported by the corresponding P-values, which are extremely low for all columns.
The Phillips-Perron test statistic is also highly negative, confirming stationarity for all indices.These statistics are generally lower than the ADF statistics, which is consistent with the tendency of the Phillips-Perron test to produce more negative values.
The results of both tests converge, indicating that discretization successfully made all columns stationary.This means that trends and seasonality in the data have been removed and the data is now suitable for time series analysis.High absolute values of ADF and Phillips-Perron statistics in all columns indicate a robust degree of stationarity.This is particularly important for financial time series analysis as it allows for more robust modeling and forecasting (Wang, 2022).

Correlation analysis
The calculated Pearson correlation coefficients (Table 3) are unconditional, that is, they measure the linear relationship between pairs of indices without considering external factors or time dependencies.They represent a simple linear association between indices.There are a few cases of negative correlations, but they are generally weak.Notably, SSEC has a slight negative correlation with FTSE and BVSP.Some indices, such as SSEC and BVSP, show relatively low correlations with most other indices, suggesting weaker linear relationships.

Static general volatility correlation
When looking at the static total volatility correlation (Table 4), it is important to pay attention to the "From" values, which are given in the far right column of Table 4 ("From others" column).
The "From" variables represent the effect of volatility transferred from the other 11 variables to each individual variable in the context of the forecast error variance.The "From" values range from 74.93% (FCHI) to 3.19% (CBDCAI).Most "From" For the static net directional connectedness of total volatility spillover, which is also presented in Table 4 (row "Net directional Connectedness"), the net values indicate the difference between the static connection of total volatility spillover with others and the connection of static total volatility spillover from others.The "net" value for CBDCAI was negative at -1.06%, indicating that the impact of CBDCAI on the volatility of the other 11 variables is smaller than the impact of volatility of the other 11 variables on it.In other words, CBDCAI absorbs the impact of other indices, but has a limited effect on their volatility.Therefore, in stressful situations, CBDCAI can act as a spillover buffer.In comparison, FTSE is the most important source of volatility with 205.02% of total volatility.
The off-diagonal elements of the 12 x 12 matrix in Table 4 illustrate the static "Net" pairwise volatility relationship between the volatilities of the two variables.For example, the value 0.30% in row 5, column 2, represents the percentage of the variance expansion of the forecast error of FTSE volatility due to shocks from CBDCAI.Regarding the impact from CBDCAI, the correlation of static "Net" pairwise volatility between CBDCAI and other financial markets is low, ranging from 0.02% (NSEI) to 0.84% (SSEC).Most of CBDCAI's volatility is the result of domestic shocks (96.81%).

Dynamic spread of total volatility
The previous empirical analysis showed static connectivity based on the full sample of data.However, it is important to consider how this relationship may change over time in low-frequency data and how it may reflect the dynamic relationship between CBDCAI and other stock market indices.To do this, this section will determine the dynamic dependence of the spillover of total volatility using fixed sample data.
Figure 2 shows how the assessment of the spillover effect changes over time.As indicated above and shown in Table 5, the static overall connectivity index is 48.91%.However, the actual total connectivity index (TCI) varies from 54.35% to 87.21% (Figure 2).That is why it is so important to take a close look at the TCI timeframe.CBDC timing methodology can also provide investors, stakeholders, and policymakers with useful information on volatility spillovers.This indicator reflects the degree of interrelationship for the entire system of variables.Using TVP-VAR, the figure illustrates how the relationship between volatility variables changes dynamically over time.The baseline variance schedule assumes a 52-week horizon (H = 52) for the sample period from January 2015 to July 2023.
In periods 2018-2019 and 2020-2021, a number of important events occurred that significantly affected the relationship between CBDC and stock exchange indices.Therefore, in the periods of 2018-2019 and 2020-2021, the world of CBDC and the stock markets were strongly connected due to global financial and economic challenges, as well as the growing interest in digital central bank currencies.index.However, in 2017 CBDCAI changed from a volatility absorber to a volatility transmitter.This may be due to the global discussions regarding the implementation of CBDC.But as shown in Figure 3, usually CBDC acts as a volatility absorber.

Net pairwise volatility spillover
The values in Figure 4 are "net" pairwise directional volatility spillovers that represent the difference between the relationship of the static total directional volatility spillover to others and the relationship of the static total directional volatility spillover from others.
Variations in the net directional effects of volatility are observed over the covered period.It should also be noted that the data reflect the dynamic nature of financial markets.For example, there are negative spillovers from most indices to CBDCAI in the initial period.However, this pattern changes over time and some indices become sources of volatility.
CBDCAI's net volatility values highlight the degree of correlation with traditional stock markets.It has become more interconnected with certain markets over time, while becoming disconnected from others.
The analysis of this data can help investors and financial analysts assess the risks and potential effects of the connection between the CBDC market and traditional stock markets.Positive values can mean that shocks in the CBDC market can affect traditional markets, and vice versa.

CONCLUSIONS
Several key conclusions can be made based on the analysis.First, the interdependence and mutual influence between the digital currency market of central banks and the stock markets was revealed.Second, even with this interaction, the CBDC market typically operates relatively independently from other stock markets.Third, CBDCAI assumes the influence of other indices, but has a limited effect on their volatility.Over time, CBDCAI has been seen to become more interconnected with certain markets while becoming disconnected from others.Fourth, despite the general ability of the CBDC market to absorb fluctuations in volatility, it may also change its function with the widespread adoption of central bank digital currencies in many countries.
The analysis of the resulting data can help investors and financial analysts to assess the risks and potential effects of the connection between the CBDC market and traditional stock markets.Moreover, investors can use the obtained results to improve risk management and optimize their investment portfolios.
In addition, for regulators and policymakers, this document also offers several advantages.For example, it can help them get more information on the impact of CBDC news on stock markets.
There are several limitations to the study.First, because CBDCs are still in their early stages of development, the availability of data is limited.Second, the analysis is limited to the impact of CBDC news on the stock market only.The final results will become more accurate once CBDC is widely implemented.Third, much of the existing research on CBDC is theoretical, making it difficult to compare the results of this study with previous studies.Finally, once the majority of central banks have adopted CBDC, it will be possible to conduct additional research to assess the impact of CBDC on monetary policy mechanisms and various financial markets.Therefore, this topic remains relevant for future scientific research.
It should be noted that in the future, we could also explore the relationship of CBDCAI volatility with the volatility of cryptocurrencies, fiat currencies, or try using different econometric models.
Note: Vertically: the value of the natural logarithm of the index value.

Figure 1 .
Figure 1.Dataset of input data j th ε t to the h-step variance of the forecast error of variable k.

FCHI
and SPX indices show a strong positive correlation of around 0.890, indicating that these two financial indicators tend to move positively together.AEX and SSMI show a strong positive correlation of approximately 0.944, indicating a similar movement pattern.N225 and NSEI also show a strong positive correlation of approximately 0.957.CBDCAI and FCHI have a moderate positive correlation of approximately 0.730.SPX and DAX show a moderate positive correlation of approximately 0.553.KOSPI and FTSE show a moderate positive correlation of approximately 0.382.
The peak was in 2018-2019: During these years the interest in CBDCs increased dramatically and many central banks began researching the possibility of implementing their own digital currencies.This has led to discussions and analysis in the global financial environment.Stock indices have become highly correlated as world events and news strongly influence stock markets.Various aspects of the economy and geopolitical events have affected stock market indices, and this has led to fluctuations and correlations between them.Decisions by central banks to implement CBDCs had the potential to bring major changes to the financial services and investment landscape.The peak of 2020-2021: This period was marked by the global COVID-19 pandemic, which caused severe turbulence in global stock markets.Stock indices have suffered heavy losses due to the spread of the virus and the subsequent measures.But even after the recession, stock markets responded strongly to stimulus measures introduced by central banks and governments to keep the economy afloat.This has also affected the interest in CBDCs.Note: Vertically: the percentage value of connectedness shows the degree of interconnectedness for the entire system of variables.

Figure 2 .
Figure 2. Total volatility of the system

Figure 3
Figure 3 shows the relationship in volatility transmission, showing how volatility spreads among markets.The vast majority of volatility values for CBDCAI are negative, indicating that volatility is transmitted to the CBDC market more than

Figure 3 Figure 3 .
Figure3shows that FTSE, NSEI, SSEC and BVSP are key sources of volatility in stock markets.On the other hand, SSEC and N225 appear to be volatility absorbers, similar to the CBDC attention

Note:
Negative values mean that in the specified pair CBDCAI receives directional volatility spillovers, positive values -it is a source of directional volatility spillovers.

Figure 4 .
Figure 4. Pairwise directional volatility spillover of the CBDCAI index Grothoff et al. (2021)ed that a well-regulated CBDC seems possible, but this does not mean that CBDC will not become a catalyst for change in the financial system.Lee et al. (2020)concluded that CBDC will be a major tool in the future digital economy, and countries familiar with the technology will have a competitive advantage.willreduceprivacyrisksand increase public confidence in the retail CBDC system.Bijlsma et al. (2021)suggests that central banks can drive consumer acceptance of CBDC through interest rates, a way to implement CBDC that considers society's needs for security and privacy and a clear message about what CBDC entails.Grothoff et al. (2021)conclude that a privacy-preserving and regulatory-compliant way of implementing CBDC can offer an appropriate balance between transaction privacy and legal compliance.Rybski (2023) recommends that central banks emphasize sustainability in their strategy.
Sethaput and Innet (2023)s at the expense of the destruction of deposit base for commercial banks.operability.Also,Sethaput and Innet (2023)claim that blockchain or distributed ledger technology can be used to implement CBDC in conducting and accounting for peer-to-peer transactions.With the emergence of private money such as cryptocurrencies and stablecoins, as well as the growing use of digital payments to combat the global spread of pandemic, CBDC is becoming an active area of research for central banks around the world.Many of them have started their CBDC projects by building trial models based on distributed ledgers (PoCs) to scale up wholesale payment systems and explore possible applications such as delivery of goods (DvP) and international money transfers.Countries with large economies, such as the United States of America, are also actively exploring CBDC issues.For example, the People's Bank of China (PBoC) is already piloting its digital currency for retail payments.Central bank digital currency transaction accounting systems use blockchain to exchange information with the central bank.Public data on currency exchange between countries can provide information on the flow of money between those countries.The paper proposes a blockchain system and management method based on the ISO/IEC 11179 metadata exchange standard to exchange information between registered CBDCs and to record transactions between them.One group of research concerns the security and privacy of CBDC implementation, conducting and tracking transactions, etc. Atako (2021) argues that a proper privacy system governing re-tail CBDC records . It also focuses on the world's most influential indices, which reflect global stock markets and the world's largest economies.For example, Wang et al. (2022) examine the relationship between CBDC and the global stock market, in particular the FTSE All World index.Similarly, Van der Westhuizen et al. (2022) use the MSCI All World Index to predict volatility in global markets.
(Wang et al., 2022)stochastic volatility developed by Primiceri in 2005(Primiceri, 2005)was used.This model was applied to weekly data from January 2015 to August 2023 to determine the impact of a central bank digital currency attention index on stock markets.The CBDCAI index, which was developed in 2022(Wang et al., 2022), covers over 660 million news sources from LexisNexis News and Business that have been available since January 2015, when Ecuador first implemented its own CBDC.
, CBDCAI t denotes the natural logarithm of CBDC news attention index calculated by Wang et al. (2022); FCHI t is the natural logarithm of the French stock market index, SPX t is the natural logarithm of the US stock market index, DAX t is the natural logarithm of the German stock market, FTSE t is the natural logarithm of the UK stock market, SSEC t is the natural logarithm of the Chinese stock market, AEX t is the natural logarithm of the stock market in Netherlands, SSMI t is the natural logarithm of the Swiss stock market, N225 t is the natural logarithm of the Japanese stock market, NSEI t is the natural logarithm of the stock market in India, BVSP t is the natural logarithm of the Brazilian stock market, KOSPI t is the natural logarithm of the South Korean stock market. where Banks and Bank Systems, Volume 18, Issue 4, 2023 http://dx.doi.org/10.21511/bbs.18(4).2023.14

Table 1 .
Descriptive data analysis

Table 2 .
ADF and Philips-Perron stationarity tests

Table 4 .
Statistical distribution of volatility The values show some fluctuations, indicating changing levels of correlation between CBDCAI and other indices.Negative values in the table indicate that CBDCAI tends to receive more directional volatility spillovers from other indices, while positive values indicate that CBDCAI is