- Jul 08, 2025
- Strategy
In the age of information explosion, the formation and evolution of market expectations increasingly rely on the in-depth analysis of unstructured data. HSAM's Macro Strategy (MS) centers on unstructured data processing, establishing a market expectation management system comprising “information mining - expectation modeling - risk warning.” This strategy breaks through the limitations of traditional technical analysis by real-time processing of non-structured data such as policy texts, industry news, and social media sentiment, accurately capturing changes in market expectations and providing investors with macro decision-making support that leads the market.
Data Architecture: Integrated Analysis of Multi-Source Non-Structured Data
The core advantage of the MS strategy lies in its ability to integrate and process five major categories of non-structured data:
Policy Text Data: Real-time capture of central bank statements, government work reports, and ministerial documents, using keyword weight analysis (e.g., the frequency and context of terms like “monetary policy” and “fiscal stimulus”) to assess marginal changes in policy direction. For example, when the phrasing of “expanding domestic demand” in quarterly policy documents shifts from “timely implementation” to “priority implementation,” the strategy interprets this as a signal of rising expectations in the consumer sector.
Industry Sentiment Data: Integrates industry media reports, expert comments, and supply chain research summaries to construct non-financial indicators of industry sentiment. For example, in the semiconductor industry, by analyzing the media exposure of keywords such as “capacity utilization rate” and “technological breakthroughs,” the strategy predicts industry cycle inflection points to complement the lagginess of traditional financial indicators.
Corporate narrative data: Deeply analyze corporate financial statement notes, investor relations Q&A, and public speeches by senior executives to uncover early signals of fundamental changes. When a senior executive of a listed company frequently mentions “expanding into overseas markets” during roadshows with detailed descriptions, the strategy will interpret this as a potential signal of corporate strategic transformation rather than superficial narrative.
Social media data: Monitor user discussions on financial forums and social platforms to construct market sentiment indicators. For example, after the release of macroeconomic data, analyze the sentiment trends of related topics (such as the proportion of “optimistic” vs. “concerned” comments) to assess the deviation between market expectations and the actual impact of the data.
Alternative data such as satellite imagery: Process satellite remote sensing data, transportation and logistics data, etc., using image recognition technology to validate the micro-level implementation of macro-level logic. For example, analyze satellite imagery of port container throughput to corroborate trends in import and export data, enhancing the authenticity of macro-level analysis.
Strategy application: Risk-return management in changing market expectations
The MS strategy converts unstructured data into three categories of investment decision support:
Capturing expectation gaps: When unstructured data reflects market expectations (such as social media interpretations of a policy) that significantly deviate from the actual policy content, the strategy identifies investment opportunities arising from these expectation gaps. For example, after the introduction of regulatory policies for a certain industry, if social media overly interprets them as negative, while the actual impact of the policy text is neutral, the strategy will position itself in related assets that have been unfairly punished.
Trend Sustainability Assessment: By tracking trends in unstructured data, the strategy assesses the sustainability of macroeconomic trends. For instance, during an uptrend in the new energy sector, if industry policy texts consistently signal positive developments, companies frequently announce capacity expansions, and social media attention remains high, the trend is deemed to have strong sustainability; conversely, it signals a potential trend reversal risk.
Tail Risk Warning: By identifying abnormal fluctuations in unstructured data, the strategy detects potential risks. When the phrase “preventing systemic risks” suddenly increases in central bank policy documents, the keyword “uncertainty” rises in industry expert comments, and unconventional panic discussions appear on social media, the strategy will activate the risk warning mechanism to reduce the portfolio's risk exposure.
Strategy Value: Pricing Expectations in Unstructured Data
The core competitiveness of the MS strategy lies in its precise grasp of market expectations. In the context of a complex and ever-changing macroeconomic environment and diverse policy interpretations, market prices not only reflect known information but also incorporate expectations about the future. Unstructured data serves as a key carrier of market expectations. For example, when a country's fiscal stimulus plan has not yet been officially announced, related policy discussions may already be circulating in industry media and social media. The MS strategy analyzes this unstructured data to anticipate market expectations regarding the stimulus's magnitude and adjust asset allocation accordingly. This ability to price expectations based on unstructured data enables the strategy to position itself before macro trends form, reap returns when market consensus is established, and avoid risks when expectations are overly inflated, providing investors with a macro risk management tool adapted to complex market environments.