4.1. Static quantile connectedness
To examine the high correlation of traditional assets or AI super powered Funds and coins (Hasselgren et al., 2024), we apply a variant of the Die Bold and Yilmaz 2013 means based metrics, namely the quantile based self- regressive vectors (QVAR) of He et al. By using it, it is possible to observe the interconnection of the values reflected in the parameters of the method in various financial conditions, for instance, a market with an indicator equal to zero as an average measure. 5, a very negative marketplace with a measure only at level zero. 05, while the measure of the market was found to be extremely favorable with a very high value of 0. 95. Table 3 shows the estimated values of the static QVAR method (Eq. 1) and connectivity metrics, which are the total connections, index (TCI), and directions closeness (TO, VIA, and NET).
Table 3
spillovers at the mean (q = 0.60) that are structural.
| SEA | FET | AGIX | ROBT | IRBO | ROBO | Golden | BTC | Fat | Fairness | REIT | Promise | USDI | FROM |
SEA | 46.48 | 22.26 | 8.20 | 4.40 | 4.48 | 4.60 | 2.42 | 22.20 | 2.24 | 4.84 | 2.60 | 2.04 | 2.42 | 64.64 |
FET | 22.80 | 42.88 | 0.22 | 4.04 | 4.00 | 4.28 | 2.66 | 22.42 | 0.08 | 4.06 | 2.06 | 2.40 | 2.68 | 68.22 |
AGIX | 8.66 | 8.40 | 62.42 | 4.42 | 4.60 | 4.84 | 2.24 | 20.40 | 2.08 | 4.04 | 2.82 | 2.24 | 2.00 | 48.68 |
ROBT | 2.20 | 2.48 | 2.26 | 22.22 | 28.64 | 28.24 | 2.00 | 4.80 | 2.64 | 26.48 | 6.02 | 2.00 | 2.88 | 88.80 |
IRBO | 2.00 | 2.20 | 2.46 | 28.42 | 22.02 | 28.84 | 2.22 | 4.08 | 2.46 | 24.82 | 6.82 | 2.82 | 2.08 | 88.08 |
ROBO | 2.08 | 2.46 | 2.26 | 28.88 | 26.40 | 22.20 | 2.20 | 4.82 | 2.60 | 26.64 | 8.60 | 2.80 | 4.46 | 88.82 |
Golden | 2.86 | 2.86 | 2.06 | 4.28 | 4.26 | 4.42 | 40.68 | 4.60 | 4.04 | 4.46 | 4.24 | 8.26 | 0.60 | 60.44 |
BTC | 20.24 | 0.68 | 0.28 | 6.20 | 6.84 | 6.64 | 2.42 | 48.08 | 2.02 | 6.02 | 4.20 | 2.42 | 2.42 | 62.02 |
Fat | 2.66 | 2.42 | 2.26 | 4.64 | 4.62 | 4.04 | 4.26 | 2.28 | 66.24 | 4.86 | 2.40 | 2.24 | 2.66 | 44.88 |
Fairness | 2.02 | 2.22 | 2.42 | 26.62 | 24.08 | 26.84 | 2.44 | 4.80 | 2.28 | 22.66 | 20.22 | 2.66 | 4.42 | 88.44 |
REIT | 2.62 | 2.40 | 2.60 | 0.02 | 8.84 | 20.88 | 2.60 | 4.20 | 2.62 | 24.04 | 44.82 | 4.86 | 2.86 | 66.20 |
Promise | 2.26 | 2.88 | 2.42 | 6.00 | 6.24 | 6.28 | 8.46 | 2.22 | 2.42 | 6.06 | 4.02 | 62.26 | 4.06 | 48.86 |
USDI | 2.86 | 2.40 | 2.02 | 8.84 | 6.86 | 8.80 | 8.80 | 4.64 | 2.46 | 8.46 | 4.08 | 4.00 | 48.22 | 62.80 |
NEAR | 46.02 | 40.48 | 46.00 | 00.48 | 04.40 | 202.44 | 48.68 | 64.40 | 22.62 | 202.06 | 68.44 | 44.28 | 48.06 | 804.60 |
Inc. Own | 02.40 | 02.24 | 08.22 | 222.68 | 226.42 | 224.64 | 88.24 | 202.28 | 86.64 | 224.82 | 02.26 | 86.42 | 86.28 | TCI |
LEFT | −8.62 | −8.86 | −2.88 | 22.68 | 26.42 | 24.64 | −22.86 | 2.28 | −24.46 | 24.82 | −8.86 | −24.68 | | |
Using the QVAR connection approach in a normal market situation that was earlier described, the following results have been obtained Under the condition when q = 0. 5, Table 3 presents the means & quantile for all measurements at HI combine and I-404. Analyzing the directional FROM column of the results, it was identified that the market spillovers are most significant in the stock market and AI ETFs. In particular, the nature of segmentation is such that there is almost a 78% spillover between the stock market and Not one, not two, but four AI funds. On the other hand, the bond and energy markets indicate almost no case of systemic spillovers (X. Zhang et al., 2022). There is a visible trend here in the data; energy markets are the most addressed at 34.87% of all spillovers, and bond markets globally acquired spillovers. 47.75%. Based on the data, these markets have no relation with AL ETFs and other related markets referred to as standard markets. However, AI ETFs face more substantial system spills than AI coins on the blockchain (Fig. 3).
More specifically, the findings indicate that the stock market and AI ETFs have the most impact on other system factors. The very high spillover values—102% for ROBO and the stock market combined—and the relatively low values—99% and 93% for ROBT and IRBO, respectively, make this point clear. Our research shows that bonds (34.17% of the total) and gold (38.57%) have the most negligible impact on the system. This supports the theory That the oil, bond, and metal prices have little to no impact on AI coins and Iris (Fig. 4).
The Fig. 5 demonstrates the substantial Function of Nan exchange-traded funds in guiding systems and doing various types of price forecasting. A lower level of system connectivity is also shown by the mean number of emitted (received) spillovers, which range from 48–57% for AI tokens. Because of their strong correlation and decentralization, our findings emphasize the value of AI tokens as a diversifier for conventional asset portfolios. According to the NET row's analysis of net conditional connection, AI Etf and cryptocurrency sales have spillover impacts on other assets (Kumar & Singh, 2022). The positive outcomes of the unconditional NET measure clearly show this. However, the low numbers of the NET measures indicate a net inflow of spillovers into the AI token market and the other markets under consideration. Contrary to expectations, the median quantile of spillover for AI tokens is the lowest, indicating little to no relationship between AI coins and other markets.
Even under typical conditions, the high TCI value (61.04%) in Table 3 indicates a highly interdependent system. Diebold and Yilmaz's (2012, 2014) mean-based connection approach will be our strength and comparison capability benchmark. We will compare this approach to the quantile connection technique at the median quantile. In the appendix, you may find the findings in Table A1. able A1 in the appendix shows the results obtained via the DY mean-based method. Results produced with the static DY connections method are similar to those using the median quantile for several connection metrics. In terms of global connectivity, the value is 62.25%., conditional connection (both ways) to and from Measures a system's net direction connectivity. (Atri et al., 2021). are among the researchers whose work is consistent with these results. Figure 6 compares median connection with mean-based connectivity, paying close attention to the two extremes at the top and bottom of the distribution. The findings presented here are derived from studies conducted by (EryiĞit, 2017). We can differentiate between large negative and negative shocks (q = 0.95) and q = 0.05, using the connectivity measures at the highest and lowest quantiles. For the extreme upper quantile relationship, see Table 4; for the severe upper quantile connection, see Table 5. All markets considered have the same systemic spillover effects when the circumstances are unfavorable, including at the lowest quantiles (q = 0.05). Evidence of this may.be seen in the system's FROM column. Tokens backed by artificial intelligence, exchange-traded funds, and traditional assets are all vulnerable to significant disruptions in the system.
Table 4
The very lowest measure (q = 0.06) exhibits still ripples.
| DEEP | FET | AGIX | ROBT | IRBO | ROBO | Golden | BTC | Grease | Justness | REIT | Promise | USDI | FROM |
DEEP | 8.08 | 8.62 | 6.44 | 8.68 | 8.66 | 8.82 | 8.42 | 8.00 | 8.42 | 8.22 | 8.42 | 8.06 | 8.42 | 02.02 |
FET | 8.24 | 8.62 | 6.68 | 8.86 | 8.66 | 8.80 | 8.06 | 8.02 | 8.46 | 8.24 | 8.40 | 6.04 | 8.42 | 02.48 |
AGIX | 8.02 | 8.80 | 8.64 | 8.62 | 8.60 | 8.82 | 8.08 | 8.08 | 8.22 | 8.22 | 8.42 | 6.88 | 8.46 | 02.48 |
ROBT | 6.60 | 8.44 | 6.28 | 8.66 | 8.20 | 8.66 | 8.06 | 8.44 | 8.44 | 8.82 | 8.66 | 8.20 | 8.28 | 02.44 |
IRBO | 6.48 | 8.20 | 6.28 | 8.46 | 8.60 | 8.66 | 8.00 | 8.60 | 8.68 | 8.84 | 8.60 | 6.08 | 8.26 | 02.42 |
ROBO | 6.42 | 8.24 | 6.48 | 8.44 | 8.40 | 8.80 | 8.22 | 8.40 | 8.62 | 8.80 | 8.66 | 8.24 | 6.06 | 02.22 |
Golden | 6.62 | 8.24 | 6.24 | 8.86 | 8.48 | 8.68 | 20.24 | 8.60 | 8.82 | 8.20 | 8.60 | 8.66 | 8.62 | 80.88 |
BTC | 6.08 | 8.80 | 6.60 | 8.00 | 8.86 | 8.02 | 8.28 | 8.82 | 8.40 | 8.06 | 8.42 | 6.04 | 8.48 | 02.20 |
Grease | 6.48 | 8.04 | 6.26 | 8.82 | 8.62 | 8.62 | 8.60 | 8.48 | 0.80 | 8.20 | 8.42 | 8.28 | 8.26 | 00.40 |
Justice | 6.66 | 8.24 | 6.28 | 8.24 | 8.80 | 8.60 | 8.28 | 8.64 | 8.20 | 0.28 | 8.84 | 8.00 | 8.40 | 00.84 |
REIT | 6.68 | 8.66 | 6.28 | 8.60 | 8.84 | 8.06 | 8.20 | 8.64 | 8.48 | 8.60 | 0.84 | 8.24 | 8.46 | 00.28 |
Promise | 6.66 | 8.08 | 6.08 | 8.04 | 8.60 | 8.00 | 8.66 | 8.40 | 8.64 | 8.46 | 8.64 | 8.66 | 8.64 | 02.44 |
USDI | 6.82 | 8.40 | 6.44 | 8.66 | 8.46 | 8.68 | 8.48 | 8.48 | 8.82 | 8.22 | 8.44 | 8.26 | 0.62 | 00.40 |
NEAR | 80.00 | 88.08 | 86.88 | 04.82 | 02.64 | 06.84 | 08.82 | 00.84 | 80.02 | 200.40 | 202.08 | 86.22 | 88.82 | 2282.84 |
Inc.Private | 88.08 | 06.80 | 84.42 | 204.28 | 202.24 | 204.84 | 208.04 | 00.64 | 00.62 | 200.68 | 222.82 | 04.88 | 08.42 | TCI |
LEFT | −22.02 | −4.40 | −26.60 | 4.28 | 2.24 | 4.84 | 8.04 | −0.46 | −0.40 | 0.68 | 22.82 | | | |
Table 5
A severe top measure (q = 0.06) exhibits stable ripples.
| DEEP | FET | AGIX | ROBT | IRBO | ROBO | Golden | BTC | Oil | Equity | REIT | Bond | USDI | FROM |
SEA | 8.82 | 8.64 | 8.62 | 8.64 | 8.22 | 8.88 | 8.26 | 8.06 | 8.22 | 8.64 | 8.64 | 8.02 | 8.20 | 02.20 |
FET | 8.80 | 0.28 | 8.68 | 8.66 | 8.22 | 8.86 | 8.20 | 8.84 | 8.06 | 8.46 | 8.60 | 8.00 | 8.00 | 00.82 |
AGIX | 8.64 | 8.60 | 20.64 | 8.24 | 6.00 | 8.60 | 8.22 | 8.64 | 8.20 | 8.44 | 8.62 | 8.02 | 8.88 | 80.46 |
ROBT | 8.04 | 8.04 | 6.08 | 0.42 | 8.20 | 8.02 | 6.86 | 8.24 | 8.04 | 8.42 | 8.80 | 8.04 | 8.46 | 00.68 |
IRBO | 8.22 | 8.22 | 6.08 | 8.68 | 8.66 | 8.82 | 6.00 | 8.20 | 6.00 | 8.22 | 8.66 | 8.24 | 8.48 | 02.44 |
ROBO | 8.00 | 8.02 | 8.26 | 8.68 | 8.06 | 8.06 | 8.00 | 8.26 | 8.26 | 8.46 | 8.84 | 8.42 | 8.44 | 02.04 |
Golden | 8.64 | 8.24 | 8.60 | 8.46 | 8.22 | 8.66 | 0.42 | 8.48 | 8.44 | 8.48 | 8.62 | 8.68 | 8.62 | 00.68 |
BTC | 8.02 | 8.68 | 8.62 | 8.62 | 8.24 | 8.86 | 8.22 | 0.44 | 6.04 | 8.46 | 8.40 | 8.02 | 8.86 | 00.68 |
Grease | 8.24 | 8.28 | 8.04 | 8.66 | 8.20 | 8.82 | 8.22 | 8.42 | 0.00 | 8.80 | 8.84 | 8.46 | 8.80 | 00.20 |
Justice | 8.08 | 8.08 | 6.00 | 8.46 | 8.02 | 8.86 | 8.22 | 8.26 | 6.00 | 8.80 | 8.06 | 8.06 | 8.40 | 02.22 |
REIT | 8.22 | 8.26 | 8.40 | 8.80 | 8.48 | 8.24 | 8.06 | 8.46 | 6.00 | 8.06 | 0.46 | 8.28 | 8.86 | 00.64 |
Promise | 8.42 | 8.40 | 8.28 | 8.64 | 8.26 | 8.84 | 8.00 | 8.26 | 6.08 | 8.48 | 8.68 | 20.28 | 8.24 | 80.82 |
USDI | 8.40 | 8.46 | 8.46 | 8.66 | 8.26 | 8.86 | 8.24 | 8.60 | 8.24 | 8.64 | 8.66 | 8.48 | 0.04 | 00.08 |
NEAR | 88.26 | 88.44 | 88.66 | 04.02 | 88.60 | 06.68 | 86.84 | 80.20 | 84.88 | 02.68 | 02.86 | 08.80 | 02.88 | 2288.62 |
Inc.Own | 06.08 | 06.62 | 08.40 | 204.24 | 08.26 | 206.64 | 06.26 | 08.62 | 04.68 | 202.66 | 202.22 | 208.08 | 202.82 | TCI |
LEFT | −4.04 | −4.40 | −2.80 | 4.24 | −2.86 | 6.64 | −4.84 | −2.48 | −6.44 | 2.66 | 2.22 | | | |
On the one hand, the TO column's stated numbers show that the equities, a REIT, and Artificial EEFT’s areas all contribute significantly to the economy amid major recessions, highlighting the importance of these sectors in anticipating when various markets will plummet. The dependent connectivity measurements of the Nets are reported in the final row. The beneficial Positive measurements show that markets for artificial intelligence exchange-traded funds (ETFs), gold, equities, and real estate investment trusts (REITs) emit net shocks to other systems. As a whole, the system's shocks are felt most by AI coins and other related exchanges (R. Chen & Xu, 2019). Table 5 shows that beneficial changes to the economy equally influence all sectors as they get comparable quantities of spillover at high upper percentiles (q = 0.95). Good surprises are often transmitted via AI ETFs, equities, real estate investment trusts (REITs), bonds, and currencies. Under extreme shocks, at the top percentiles, the NET contingent connectivity results reveal that the following markets serve as net shock sources: bonds, equities, real estate investment trusts (REIT), AI ETFs (except IRBO), and USD currency exchanges. Surprisingly, markets for artificial intelligence coins, gold, vitality, and cryptocurrencies are the ones that absorb upsets the most (Fig. 7).
The severe measure connection outcomes show that total connection (TCI) increases significantly at both the top and bottom of the distribution exceeding 90%. The connection at the mean measure (DY unity measure) and the mean measure (q = 0.5) are only 62.25% and 61.04%, respectively. In addition, compared to the mean (DY connectivity metrics) and average measure (q = 0.5), the connection values at the extreme quantiles, which include contingent impacts FROM OR TO the system, are more significant. Several additional studies have reached similar conclusions, including (Bouri et al., 2017). According to the analysis, extreme optimism and minus shocks significantly impact AI tokens and AI ETFs.
The mean-based DY connection analysis results imply that traditional markets, including gold, energy, cryptocurrency, bonds, exchanges, and stock markets, need to be adequately diversified by AI ETF assets. The strong and positive relationships seen among AI ETFs and those markets, even under normal market conditions, as detailed (Singhal et al., 2019). and Bauer and Hoang (2021), might be the reason for this. Due to their weak link to other markets under typical market circumstances, AI tokens may mitigate the risk associated with other assets. During times of significant market fluctuations, these correlations stand out more. A market's quantity of spillovers, both received and sent out, tends to spike during market volatility, especially when the economy is experiencing steep declines. Table A1 shows the average value, and Table 3 offers the median value, so we can compare this increase to what we expect in a regular market. Research shows that artificial intelligence (AI) exchange-traded funds (ETFs) can't fully diversify the assets of conventional portfolios, particularly during recessions.
Connectivity levels vary among quantiles, as seen in Fig. 6. The relationship level between system variables increases significantly at the extreme quantiles, as seen in Fig. 6. Since shocks raise correlation coefficients across variables in proportion to their severity, very positive and highly negative shocks have comparable effects on the system. (Guan et al., 2021).At the lowest quantile (q = 0.5), the TCI level is much higher than 60%. However, the overall degree of system interconnection spikes after significant disturbances, going beyond 90% at the top and bottom quantiles. The symmetrical and balanced pattern of TCI fluctuations is seen in good and bad market conditions. (Y. J. Zhang & Wei, 2010). also came to similar findings; this trend makes sense.
The discovery that AI ETFs are a significant assets sector and positive emitter in any market and situation is essential for diversity techniques and portfolio planning. A portfolio that includes assets related to artificial intelligence might be an excellent way to protect oneself from potential losses in connection with stock, property, oil, gold, and cryptocurrency markets. In addition, (Cui et al., 2023). suggest that a shift in the economy terms of AI ETFs could be a precursor to changes in the other market sectors that were part of this study. They also note that investing in fresh AI resources could be an excellent way to gauge the impact of creativity driven by problems. To get a good look at risk diversity, it is essential to look at connections at different degrees of economic trouble, including severe tail-ends, as net receivers and emitters differ across the quantiles. This is especially true for assets linked to the internet, which are subject to the most significant levels of instability and volatility in this developing market because of the fast and constant technical development and upheaval.
Prior work (W. Huang & Wu, 2021) has shown that tail-related risk might appropriately indicate system risk; this study's conclusions support that idea. When studying market connectivity, it is essential to incorporate tail-end hazards. Therefore, the tail-end distributions of impacts caused by severe either-or shocks will go unnoticed if conditioned mean-based connectivity measures or the median of dispersion are the only metrics considered. Furthermore, there is a difference between the nature of dynamic impacts at the median and the extreme tails. When evaluating investments and reducing risk, risk management and buyers should consider the unique qualities and interdependencies of traditional assets and commodities associated with artificial intelligence. Investors may precisely build diverse portfolios and discover prominent signs pointing to changes in market circumstances by recognizing the interconnection and ripple tendencies across various resources.
4.2. Dynamic quantile connectedness
According to prior empirical research, connections among Many financial developments, global issues, and recessions may cause securities to fluctuate over duration and make them insecure. The quantile connectedness approach will analyze the links among wealth, including AI ETFs or signs, plus normal. We will examine the correlations between these variables, paying close attention to the median and the top and lower quantiles in light of various economic crises and events that transpired during the given time frame. Our sample period encompasses the beginning of the COVID-19 problem and the times immediately following the commencement of vaccination campaigns.
Figure 9 shows how the total connectivity index (TCI) changes at various quantiles. For the mean quantile (q = 0.5), the blue line shows how the TCI varies near the midpoint of the distribution (Elie et al., 2019). Under typical market conditions, the degree of connectivity fluctuates between 45 and 80%. For example, just around the beginning of the COVID-19 pandemic in the first quarter of 2021, Eighty percent of TCI was reached. Afterward, the following year had a dramatic decline, even getting 50% in that last year. As the COVID-19 vaccination was widely disseminated and used, the financial markets started to seem more normal again, reducing the extreme connections among investments to more typical levels. The relationships between system variables vary within normal ranges during this time, reaching a second high of more than 75% in the fourth quarter of 2022. Lines in green and orange show a dynamic relationship in the top and bottom tails. Over the complete observation period, the system's connectivity level varies somewhat, ranging from 85–92%. This shows that the system variables are always highly interdependent, regardless of whether the market is falling or rising. This exemplifies how traditional assets, AI ETFs, and tokens are susceptible to good and bad shocks. We adopt the Quantile-Frequency Connection (M. L. Liu et al., 2013).to ensure that the Total A Connection Index (TCI) remains valid after using the quantile connectivity technique. Using this approach, we can determine the total price of Investing (TCI) for different quantiles at various rates (short, medium, and long term).
Number five is the value. As a result of using the quantile connectivity method, TCI-Total provides a thorough evaluation of connection. Using the quantile-frequency connectivity method, the TCI-long term, TCI-medium term, and TCI-short duration are computed at the center, extreme upper measure, and extreme lower quantiles, respectively. Frequency varies across several quantiles (median, higher, and lower) and time intervals (overall, short, medium, and long term). Because of this, we tell our customers to change their portfolios based on how long they want to keep their money in the market (Jain & Ghosh, 2013). One possible explanation for the discrepancies in the short-, medium-, and long-term frequencies across all quantiles is that the quantile-frequency connection model and the quantile connection model vary in their underlying assumptions. Without taking time frequency into account, our quantile connectivity model evaluates changes in connection across different quantiles. However, quantiles and temporal frequency are both included in the quantile-frequency model. The changing net directional connection at various quantiles is shown in Figs. 5, 6, and 7, respectively. These quantiles are the median (q = 0.5), the lower (q = 0.05), and the more significant (q = 0.95). According to the available data, the degree to which assets are interdependent varies and changes throughout time (Atri et al., 2021). Throughout the study period and under varying market situations, the roles of AI ETFs, AI tokens, and traditional assets as receivers or senders of spillovers vary.
During most of the research, we found that all AI ETFs were net emitters of value impacts to the system, especially under normal price conditions at average quantiles. This highlights the significant role of the new electronic securities in driving and anticipating other markets. Under severe market circumstances, all assets move from net emitter to net recipients of returns ripples in both the left and right tails. This includes AI tokens along with all adjacent markets. Typically, these assets behave as net receivers of impacts from another market. Additionally, the degree of profit carryover is much more significant in severe market situations compared to typical market conditions (Bouri et al., 2017). Because net direction connectivity changes over time, traders, politicians, and investors should keep an eye on net spillage trends to swiftly modify their investments and hedge against optimistic and bearish trends in the case of severe occurrences.
The bilateral impacts at the center, severe lower and excessive higher quantiles of a network are shown in Fig. 10. At the lowest quantile (q = 0.5), the pairwise network reveals that stock and all AI ETFs are the most critical parts of the system since they release the most impact on other sectors. In addition, conventional markets and AI tokens end up as net recipients of spillovers. The prices for bonds and USD exchanges (AI coins) are the ones that stand to gain the most (least) from consequences. The fact that gold, energy, and cryptocurrencies are the least affected by transference indicates that these markets are relatively unconnected to the structure when exchanges usually function.
The net pathways undergo a slight transformation in the face of very adverse market circumstances at the very bottom tails. When we contrast AI ETFs to other asset classes, we show that gold, equities, and real estate investment trusts (REITs) send the most impact into the system. Every market is a net recipient of impacts; the ones with the most vital connections to the ecosystem are securities and AI signs, while energy and coin markets remain weak links. Pairs share spillover analysis also reveals that swap markets, fairness, securities, REITs, AI ETFs (excluding IRBO), and AI ETFs are all net sources of spillovers under favorable markets at significantly lower percentiles (q = 0.95). Bonds and AI ETFs are the most significant participants in the method's impacts. Transfers from the structure mainly benefit every other market, including AI tokens (Mensi et al., 2017). As shown in a relative tail dependency (RTD) is good when the value is for the correct tail and negative when it is for the left tail. The findings show that RTD changes with time and becomes low in 2020, the second part of 2022, and 2022. The majority of the material collected time, nevertheless, shows adverse RTD. It shows that the left tail is more dependent on the reverse tail, which means that connections are less fragile.
Over many periods (short-term, medium-term, and long-term), this study examines the dynamic total connectivity index (TCI) for different quantiles (Cui et al., 2023). The Median quantile total connection index, which fluctuates, is shown in Panel A. In Panel B, we can see the lowest 5th percentile total connection score as a function of time (q = 0.05). At the most extreme quantile (q = 0.95), Panel C displays the overall connection index as it varies over time (Fig. 11).
Finally, the research results on dynamic quantile connectedness indicate that, on average, the degree of dynamic total connection is higher at the top and bottom quantiles than in the center. During extreme market volatility, there is a high link between AI-driven assets and conventional markets. Furthermore, the highest and lowest quantiles of the overall connectedness score exhibit variable oscillations. An asymmetrical relationship may exist since the correlation is rather persistent over time and quantiles. How these markets respond to unusual occurrences is contingent upon the kind and timing of the shocks, and AI and traditional assets react differently to positive and negative shocks, respectively (R. Chen & Xu, 2019). We find that across different quantiles and periods, AI and traditional assets both show a variation between net emitters and net recipients of return spillover. In addition, the research reveals that dynamic network connection indicators are more volatile during extreme market events compared to normal market conditions (Fig. 12). The volatility of these indicators is probably due to factors beyond our control, such as economic cycles and geopolitical events (Xu et al., 2023).Because of the expected changes in net spillover patterns throughout time, investors and portfolio managers should rebalance their assets(Dutta et al., 2020)
We may summarize the study's key points about static and dynamic connections. Many assets, such as AI tokens, AI exchange-traded funds (ETFs), gold, Bitcoin, equities, and bonds, are interconnected during severe events. Portfolio asset allocations are known to undergo significant shifts during market volatility, which aligns with the results of earlier research (Adekoya et al., 2022; Tiwari et al., 2022). When contrasted with the intermediate quantiles, there is a more unambiguous indication of disruption spread at the top quantiles. During times of crisis, the advantages of diversifying a portfolio using AI tokens and AI ETFs will likely be significantly reduced. Our study shows that AI exchange-traded funds (ETFs) are a powerful tool for system direction, and they are the best at forecasting the returns of traditional financial markets.
Furthermore, in normal market conditions, AI tokens and exchange-traded funds (ETFs) are reliable hedging vehicles for portfolios that include gold, Bitcoin, oil, and bonds. However, during economic downturns, their capacity to reduce risk diminished. Investors may use these innovative assets for hedging and diversification purposes, considering the results of (Kang et al., 2023). our study aims to improve the existing empirical research on the risk management of AI assets. We discover that AI tokens, AI ETFs, and other significant assets are highly related, which aligns with prior studies. In addition, our results are consistent with other research suggesting that during heavy market volatility, this link got larger at the extremes of the distribution. Our research shows that, excluding extreme cases, there are very few relationships among AI tokens, AI ETFs, and Bitcoin. We have yet to determine how this discovery will affect our efforts to diversify and hedge. Our study's findings suggest that the AI asset markets were unrelated to the early stages of cryptocurrency's development. The many advantages of investing across the markets for artificial intelligence assets and cryptocurrencies are supported by our empirical results, which align with (H. H. Huang et al., 2023)
In addition, disruptions in the oil, gold, and other financial markets significantly affect AI tokens, particularly during times of severe market downturn. This highlights how limited AI tokens are in their ability to provide effective hedging in emergencies. The results of our analysis show that the AI coins & CFDs provide variety and hedge benefits. change drastically depending on regular and extreme market conditions. Traditional markets and the AI business are inseparable. Therefore, investors should investigate these markets' interdependencies and transmission effects under various market conditions (Bildirici & Turkmen, 2015). In order in order to adopt well-informed investments and modify investments in response to current markets and financial conditions circumstances, it is essential to have this information.
Investors and portfolio managers may benefit from the insights we have obtained from the empirical data. During market instability, asymmetry becomes more apparent. Thus, investors should exercise caution when using AI assets to diversify their portfolios. The market for AI-generated assets is also still in its early stages. Consequently, investors must consider the market expansion potential of new assets, such as AI tokens and AI ETFs, and the risk-return analysis of asset behavior. In addition, the expanding impact of existing markets and the ever-changing patterns of development in new assets may cause changes in the structure of the financial market (Le & Chang, 2012).
4.3. Portfolio implication
Here, we calculate the ideal weight, overlay ratio, and hedged efficacy for different asset-AI coin combinations employing DCC-GARCH, and we look into client portfolio diversity. Table 6 displays the findings of this exercise, which demonstrate that most resources (IRBO, ROBO IS A, gold, capital, amid water, (MACHINE, to be optimal, the sum of (IRBO, A ROBOT, money, equality, and bond)-AGIX or (IRBO, ROBO IS A, gold, equity, and bond)-FET must be 0.99 percent. This means buyers ought to allot 99 cents of 1 dollar towards these resources and 1 cent to AI tokens (M. Arfaoui & Ben Rejeb, 2017). On the other hand, the oil-AI token combination has the lowest ideal ratio (0.85 for oil THE SEA, 0.83 for oil-EFT, plus 0.84 for oil-AGIX), suggesting that investors should put around 15 cents into the OCEAN token and 85 cents into the fuel space for every $1 invested in the oils-OCEAN portfolio.
Table 6
The ideal ratio of masses with hedges.
| Optimal weights | Hedge ratio | Hedging Effectiveness (HE) |
ROBT/DEEP | 1.08 | 1.06 | 2.06 |
IRBO/SALTY | 1.00 | 1.06 | 2.06 |
ROBO/OCEAN | 1.00 | 1.06 | 2.06 |
Golden/ABYSMAL | 1.00 | 1.02 | 2.08 |
BTC/OCEAN | 1.08 | 1.26 | 2.86 |
Grease/SEA | 1.86 | 1.06 | 2.64 |
Fairness/MARINE | 2.00 | 1.04 | 2.08 |
REIT/NAVAL | 2.08 | 1.04 | 2.06 |
Promise/SEA | 0.00 | 1.00 | 2.00 |
USDI/OCEANIC | 1.08 | 1.00 | 2.00 |
ROBT/FET | 1.00 | 1.06 | 2.06 |
IRBO/FET | 1.00 | 1.06 | 2.06 |
ROBO/FET | 1.00 | 1.06 | 2.06 |
Golden/FET | 1.08 | 1.02 | 2.08 |
BTC/FET | 1.08 | 1.26 | 2.86 |
Fat/FET | 1.84 | 1.06 | 2.88 |
Fairness/FET | 1.00 | 1.04 | 2.08 |
REIT/FET | 1.08 | 1.04 | 2.06 |
Pledge/FET | 1.00 | 1.00 | 2.00 |
USDI/FET | 1.08 | −1.02 | 2.00 |
ROBT/AGIX | 1.08 | 1.06 | .08 |
IRBO/AGIX | 1.00 | 1.06 | 2.08 |
ROBO/AGIX | 1.00 | 1.06 | 2.08 |
Gilt/AGIX | 1.00 | 1.02 | 2.00 |
BTC/AGIX | 1.06 | 1.24 | 2.04 |
Lard/AGIX | 1.84 | 1.08 | 2.06 |
Impartiality/AGIX | 1.00 | 1.14 | 1.00 |
REIT/AGIX | 1.08 | 0.24 | 1.08 |
Bond/AGIX | 1.00 | 0.12 | 1.00 |
USDI/AGIX | 1.08 | −1.02 | 1.00 |
To mitigate risk and make the most of diversity, investors should think about incorporating AI tokens into your holdings of precious metals. cryptocurrencies, stocks, bonds, real estate investment trusts (REITs), currencies, and AI exchange-traded funds (ETFs). A hedging ratio of 0.01 for gold-AI tokens suggests that a one- A cent-short AI coin option may balance out a dollar-long gold stake. An investment of $1 in AI ETF should be offset by a temporary position of $5 in AI tokens, according to the hedging ratio for AI ETF-AI tokens, which is around 0.05.
Thus, this specific type of hedge has nearly zero hedging ratio depending on the degree of risk management. In particular, starting from the 25th, the richest benefits can be seen in combining the BTC-AI token. From the avowed observations, it is discerned that AI tokens are an unreasonable means to preserve Bitcoin. However, the protection from using a combination of gold and AI tokens is the lowest; thus, gold can be hedged using AI tokens for a relatively small amount (Sajjad et al., 2022). The most successful hedging pairings are Navy connect, USDI-OCEAN, Feet connect, USDI-EFT, & AGIX USDI, the gold, equity, & bond. This means that portfolios that do not have investments in gold, fixed-income equities, or USD-related indexes may benefit the most from AI tokens.
Besides, the best pairings are the AI ETFs and AI tokens, especially AGIX. Therefore, AGIX can be a perfect diversifier that perfectly works for all the assets that were examined in this study (Soeder, 2021). The couplings with the lowest efficacy values were Equations 0.6 for oil-OCEAN, 0.76 for BTC-OCEAN, 0.78 for oil-FET, & 0.75 with BTC-FET. There is no benefit for energy and cryptocurrency market investors to diversify their portfolios with AI tokens (OCEAN and FET).