Cookie Consent by Free Privacy Policy Generator website
top of page
Search
  • Writer's pictureGeopolitics.Λsia

Weekly Trend Radar and Portfolio Investment

Updated: Jul 31, 2023

This week's Trend Radar Monitoring Section is truly exceptional. We have implemented "prompt engineering" to enhance our Generative AI's analytical report results. Moreover, we utilized the "after training" method in conjunction with the metageopolitical knowledge graph to analyze the weekly PESTLE report. We also applied the metageopolitical knowledge graph to examine the ongoing war between Russia and Ukraine. In the most thrilling segment of this week, we challenged the AI to employ its expertise in designing an investment portfolio aimed at addressing the current war situation. Financial investment serves as the most demanding testing ground for our AI model (excluding actual geopolitical conflict, which falls outside our purview as a civilian geopolitical think tank). We supplied the AI with real-time data on 10-year government bond and hedge fund index, and evaluated its performance by backtesting and readjusting the portfolio. The results were truly remarkable!




Original Portfolio (left-hand side) vs. Adjusted Portfolio (right-hand side), see details in PART III


 

Part I: PESTLE Analysis Section


This week's analysis of metageopolitical developments focuses on critical priorities in various regions, organized according to the PESTLE framework:



Weekly Trend Radar, please access the full information at http://www.geopolitics.io



Political

  • Cyber Sovereignty Pursued - Governments around the world are increasingly pursuing cyber sovereignty, which will likely result in legal and regulatory shifts. This trend showcases the importance of noopolitik and hard power dynamics in determining cybersecurity strategies.

  • Chinese Information Collection - China's efforts to collect information on key foreign individuals through a company linked to the state highlight the essential role of noopolitik elements in the geopolitical power game. Non-state actors play a vital role in amplifying state actors' efforts to build networks and engage in intelligence collection.

  • US Ukraine Support - The United States has reassured its support for Ukraine's independence, sovereignty, and territorial integrity in the face of Russian aggression. This scenario demonstrates the hard power dynamics in the region and emphasizes the strategic calculations of major state actors.


Economic

  • Bank Reforms UK - Significant bank reforms in the UK have improved the resilience and competitiveness of the financial sector. These reforms clearly exhibit the importance of economic power in shaping international relations and reflect the neoliberal metageopolitical approach to focusing on global institutions and economic factors.

  • Dollar's Global Hold - The US dollar continues to play a significant role in global financial transactions, revealing the ongoing economic power dynamics and influence.

  • BRICS Currency - A former White House adviser, Joseph Sullivan, has stated that a BRICS currency, issued by Brazil, Russia, India, China, and South Africa, could pose a significant threat to the US dollar's dominance. He argues that a BRICS currency would offer key advantages over existing alternatives, such as bilateral deals, and could achieve a level of self-sufficiency in international trade that other currency unions have not. Additionally, non-member countries may be enticed to use the BRICS currency due to the large economies of each member nation. A BRICS currency could also weaken the US dollar's status as a reserve currency, as BRICS governments could encourage their households and companies to buy assets in the new currency. While Sullivan acknowledges that the dollar's reign won't end overnight, he suggests that a BRICS currency could contribute to a multi-polar regime and a gradual erosion of the dollar's dominance. (*)


Sociocultural

  • Defining 'Woman' - A current social debate examines the definition of 'woman' and reflects sociocultural changes that have real-world implications, including potential shifts in societal norms and behavior that could influence political and economic power dynamics.


Technological

  • AI-Generated Videos - The emergence of AI-generated videos may produce new concerns over disinformation and misinformation in international relations. The use of this technology contributes to noopolitik elements affecting the geopolitical landscape.


Legal

  • Biden Expanding Asylum - President Biden's plans to expand asylum programs show potential changes in US legal frameworks for migrants and the ongoing push for international human rights law reform. Here, non-state and transnational actors play a role in advocating for policy changes.


Environmental

  • Australia Renewable Energy - Australia's ambitious renewable energy initiatives underscore the connection between economic power and environmental factors in global competition. (*)

These insights offer a comprehensive understanding of the current geopolitical landscape, enabling executives to make informed decisions in their organizations while considering the dynamic interplay of various powers in multiple domains.


 

Part II: Geopolitical Analysis Section



Subject: Metageopolitical Analysis of Russia-Ukraine Conflict – Implications for Business and Investment

Dear Executive Director,

This report presents a comprehensive metageopolitical analysis of the ongoing Russia-Ukraine conflict and its potential implications for our business and investment considerations. By integrating insights from various foreign policy theories, we aim to provide a thorough understanding of the situation and its potential evolution.

Key Aspects and Implications:

1. Military capabilities:

  • Increased regional instability may lead to disruptions in supply chains and transportation routes.

  • Potential for escalation or spillover effects could impact business operations and investments in neighboring countries.


2. Economic interests:

  • The conflict may result in economic sanctions against Russia, affecting trade relationships and investment opportunities.

  • Energy markets could face volatility due to the region's importance in global energy supplies, impacting energy-dependent industries.


3. Regional dynamics:

  • Shifts in the balance of power in Eastern Europe might lead to new geopolitical alignments, potentially creating both risks and opportunities for business and investment.

  • Rising tensions between Russia and Western countries may prompt increased military spending, influencing the defense and security sectors.


4. International institutions:

  • Global institutions, such as the United Nations and European Union, may play a role in mediating the conflict, shaping potential outcomes and business environments in the region.

  • The effectiveness of international institutions in resolving the conflict may impact global economic governance and the enforcement of international norms.


Predictions for Potential Evolution:

1. Short-term:

  • Continued volatility in energy markets, with potential disruptions to supply chains and transportation.

  • Heightened geopolitical tensions may result in additional economic sanctions and trade restrictions.

2. Medium-term:

  • Possible shifts in regional dynamics, with some countries seeking to diversify their economic partnerships and reduce reliance on Russia.

  • Increased focus on renewable energy sources and energy security, creating new business and investment opportunities.

3. Long-term:

  • The outcome of the conflict may shape the future balance of power in Eastern Europe, with potential implications for the region's political and economic stability.

  • Depending on the conflict resolution, international institutions' credibility and effectiveness may be impacted, influencing global governance and the business environment.


Business and Investment Considerations:

  1. Diversify supply chains and explore alternative transportation routes to mitigate potential disruptions.

  2. Monitor developments in energy markets and consider investments in renewable energy sources to hedge against volatility.

  3. Assess the potential impact of economic sanctions on trade relationships and investment opportunities in the region.

  4. Keep abreast of shifts in regional dynamics and geopolitical alignments, adjusting business strategies accordingly.

  5. Engage with international institutions and industry associations to stay informed about potential regulatory changes and market developments.

By considering the various aspects of the Russia-Ukraine conflict and its potential evolution, we can better prepare for the challenges and opportunities it may present for our business and investment decisions. It is essential to remain vigilant and adaptive in this rapidly changing geopolitical landscape.

Best regards, Your Geopotlical Strategist Department



 

Part III: Portfolio Investment Section



The following table outlines the suggested asset allocation, positioning, holding period, and ETH asset name (if available) for the investment portfolio based on the Russia-Ukraine conflict analysis:




Please note that the suggested asset allocation, positioning, and holding period are for general guidance only and should be tailored to your specific risk tolerance, investment objectives, and time horizon. Additionally, the realization times are based on the analysis of the Russia-Ukraine conflict and may change depending on the evolving geopolitical situation and other relevant factors. Regular monitoring and adjustment of the portfolio based on market conditions and geopolitical developments are recommended.



State 1: First round of asset allocation, portfolio #1 demonstrates the best defensive performance over the war on Ukraine (backtracking 2022 - up-to-date)


While Market CAGR (2023) = (1.0553)^(1/1) - 1 = 0.0553 or 5.53% and Weighted Risk-Free Rate = (2.95 * 12/13) + (3.60 * 1/13) = 2.9654 or 2.97%. Portfolio 1: Alpha: 8.69% - This indicates that Portfolio 1 has outperformed the market by 8.69% after adjusting for the portfolio's risk (beta). A positive alpha suggests that the portfolio manager has generated excess returns above what would be expected based on the portfolio's exposure to market risk.

Beta: 0.91 - This indicates that Portfolio 1 is less volatile than the overall market. A beta value less than 1 implies that the portfolio tends to move less than the market on average. In other words, when the market goes up, Portfolio 1 is likely to go up less than the market, and when the market goes down, Portfolio 1 is likely to go down less than the market.

Portfolio 2: Alpha: 7.80% - This indicates that Portfolio 2 has outperformed the market by 7.80% after adjusting for the portfolio's risk (beta). Similar to Portfolio 1, a positive alpha suggests that the portfolio manager has generated excess returns above what would be expected based on the portfolio's exposure to market risk.

Beta: 0.94 - This indicates that Portfolio 2 is also less volatile than the overall market but slightly more volatile than Portfolio 1. A beta value less than 1 implies that the portfolio tends to move less than the market on average. However, Portfolio 2's beta is closer to 1 compared to Portfolio 1, indicating a higher sensitivity to market movements.

Portfolio 3: Alpha: 7.54% - This indicates that Portfolio 3 has outperformed the market by 7.54% after adjusting for the portfolio's risk (beta). As with the other portfolios, a positive alpha suggests that the portfolio manager has generated excess returns above what would be expected based on the portfolio's exposure to market risk.

Beta: 0.95 - This indicates that Portfolio 3 is less volatile than the overall market but has the highest volatility among the three portfolios. A beta value less than 1 implies that the portfolio tends to move less than the market on average. However, Portfolio 3's beta is very close to 1, indicating a higher sensitivity to market movements compared to Portfolio 1 and 2.

Implications: All three portfolios have positive alphas, indicating that their respective portfolio managers have generated excess returns over what would be expected based on their exposure to market risk. This suggests effective portfolio management strategies.

In terms of risk, Portfolio 1 has the lowest beta, indicating the lowest sensitivity to market movements and the least volatility among the three portfolios. Portfolio 3 has the highest beta, indicating the highest sensitivity to market movements and the most volatility among the three portfolios. Portfolio 2 falls in between Portfolio 1 and 3 in terms of risk exposure.

Investors should consider both alpha and beta when making investment decisions, as they provide valuable information about a portfolio's performance and risk profile. Depending on an investor's risk tolerance and investment goals, they may choose a portfolio with a higher alpha and/or a lower beta, depending on their preferences for potential returns and volatility. By looking for investments with alpha > 0 and beta < 1, investors are essentially screening for investments that have the potential to outperform the market while also exhibiting lower volatility or risk. However, it's important to keep in mind that past performance is not always indicative of future results. Alpha and beta are only two of many factors to consider when evaluating investments. A comprehensive analysis should also include other factors such as the asset's fundamentals, industry trends, and economic outlook.


Please note that the performance of a portfolio should be assessed over an appropriate time horizon, considering the investor's financial goals and risk tolerance. If the negative CAGR is a short-term phenomenon, it might not be representative of the portfolio's long-term potential. Market conditions can change, and a well-managed portfolio may recover and outperform cash in the long run.




Stage 2: Benchmarking portfolio (100% US stock market, 100% 10 year Treasury, and 100% Gold), Gold portfolio seems to produce impressive safe harbor asset, therefore we will use this information to adjust in the final optimized portfolio (backtracking 2022 - up-to-date)



Stage 3: Adjusted portfolio #1's performance has been improved compared to the original one, see comparative performance data in the table below (backtracking 2022 - up-to-date)




The Sharpe Ratio, Sortino Ratio, and Market Correlation are important performance indicators for assessing the risk-adjusted returns of a portfolio. Here's a brief explanation of each and how they relate to the performance of both the initial and adjusted Portfolio #1:

  1. Sharpe Ratio: This ratio measures the risk-adjusted return of an investment by comparing the investment's excess return (return above the risk-free rate) to its standard deviation (volatility). A higher Sharpe Ratio indicates a better risk-adjusted return. In this case, the adjusted Portfolio #1 has a higher Sharpe Ratio (-0.53) than the initial Portfolio #1 (-0.71), suggesting that the adjusted portfolio has a better risk-adjusted performance.

  2. Sortino Ratio: Similar to the Sharpe Ratio, the Sortino Ratio measures risk-adjusted return, but it only considers downside risk (negative volatility) rather than overall volatility. A higher Sortino Ratio indicates better downside protection and risk-adjusted return. The adjusted Portfolio #1 has a higher Sortino Ratio (-0.68) compared to the initial Portfolio #1 (-0.90), indicating that the adjusted portfolio provides better downside protection and risk-adjusted performance.

  3. Market Correlation: This measures the degree to which a portfolio's returns move in tandem with the overall market. A correlation of 1 indicates that the portfolio moves perfectly in sync with the market, while a correlation of -1 means the portfolio moves in the opposite direction. The adjusted Portfolio #1 has a slightly higher market correlation (0.93) than the initial Portfolio #1 (0.91), suggesting that the adjusted portfolio is more closely aligned with the market's movements. However, this difference is relatively small and may not have a significant impact on overall performance.

In summary, the adjusted Portfolio #1 shows improvement in both Sharpe Ratio and Sortino Ratio, indicating a better risk-adjusted return and downside protection. The slightly higher market correlation implies a closer alignment with market movements, which may or may not be desirable depending on the investor's preferences and risk tolerance.



 

Part IV: Engineering Section



Definition: Prompt Engineering, also known as In-Context Prompting, is the process of developing methods to effectively communicate with Large Language Models (LLMs) in order to guide their behavior towards desired outcomes without updating the model weights. It is an empirical science that relies on experimentation and heuristics, as the effects of prompt engineering methods can vary significantly among different models. In the context of autoregressive language models, prompt engineering aims to improve alignment and model steerability, focusing on techniques like zero-shot and few-shot learning, instructed language modeling, and chain-of-thought (CoT) prompting to make the model more aligned with human intention and facilitate better communication.


Instructed Language Modeling (ILM) is an approach that fine-tunes a pre-trained Large Language Model (LLM) to better understand user intentions and follow instructions. This method focuses on improving the model's ability to align with human intentions and reduces the cost of communication.

To achieve this, ILM uses high-quality tuples of (task instruction, input, ground truth output) during the fine-tuning process. The task instruction provides a clear, natural language description of the task the model should perform. The input is the specific data or context for which the model needs to generate an output, and the ground truth output is the correct, or "true," output that the model should produce for the given input and instruction.

Ground truth refers to the accurate or ideal output for a specific input or set of inputs. In the context of ILM, ground truth outputs serve as a reference or target for the model to learn from during the fine-tuning process. By comparing the model's generated output with the ground truth output, the model can adjust its internal representations to better align with human expectations and instructions.

One common method for fine-tuning LLMs with instructed language modeling is Reinforcement Learning from Human Feedback (RLHF). In RLHF, human feedback is used to create a reward signal that helps guide the model's learning. By incorporating this feedback, the model can improve its ability to follow instructions and produce outputs that are more in line with human intentions, thereby enhancing its overall performance and utility.


In this regards, we applied the concept of "self reinforcement learning" to enable the AI to enhance its own "meta-prompt engineering" (combined with mixed methods in zero-shot learning, one-shot learning, few-shot learning, and chain-of-thought). This approach aimed to generate superior essay iterations with each successive attempt. The results are presented in the accompanying data and charts. The optimal outcome is found at iteration number 4, which appears unable to surpass human expertise in the relevant domain.

The art and science of "prompt engineering" involve striking a balance between "creativity" (high temperature) and "reduced hallucination" (fact constraint). However, excessive fact constraint may lead to diminishing marginal utility or even negatively impact performance. To further enhance the AI's performance, a combination of a more advanced AI model, fine-tuning techniques, real-time data, enhanced long-term memory or an amalgamation of these methods could result in better performance than currently achieved.





The optimum point is at iteration#4


 

We have successfully deployed our Trend Monitoring Radar (TMR) on the Wednesday section. We deployed a Python script to scan weekly news and asked AI to prioritize and categorize it. Additionally, we instructed AI to generate a weekly summary report based on the scanning data in a reporting style for the executive director of the Geopolitical Analysis department. The report is far from perfect, so please do not share the contents publicly. We will strive to make the entire process 100% automated and find ways to improve relevance, consistency, and accuracy in the future. our TMR is inspired by opensource.zalando.com/tech-radar/.


Geopolitics.Asia will provide serious policy analysis on Mondays, trend monitoring on weekdays, and cultural and lifestyle issues on weekends. Please note that our weekday situation monitoring will not include a trend radar or scenario analysis for the time being, as we work to fully automate these processes with AI. You can, however, access to our previous experiments on trend radar and scenario planning generated by the AI, 1) Simple scenario planning at Jan 26, 2023, 2) Double iteration scenario planning technique at February 2, 2023, 3) Triple iteration scenario planning techniqueat February 9, 2023, and 4) Hyperdimensional scenario planning technique at February 17, 2023.


Stay tuned for updates on this exciting development!




Comments


bottom of page