Market Synthesis: AI Engine for Macro Imbalances

Built for investment teams that want earlier warning of market regime breaks, with reasoning they can audit.

Most large drawdowns do not come from small forecasting errors. They come from sudden shifts in policy, liquidity, correlation, or narrative.
Market Synthesis is designed to flag those shifts before they are obvious in price alone.

The core idea: Soros-style imbalance detection

The system is built around a simple market reality: when market narrative and structural reality diverge, price can move violently when that gap closes.

This is the same style of thinking behind classic macro dislocation trades:

  1. consensus says one thing
  2. underlying constraints say another
  3. policy makers try to defend the old narrative
  4. the imbalance eventually resolves in price

Market Synthesis is designed to detect those high-tension setups early and score how actionable they are.

What this project does

Market Synthesis is a decision-support engine for macro and multi-asset teams.
Given an event context and relevant market narrative, it produces:

  1. a directional view (long, short, neutral)
  2. a confidence score (0-100)
  3. a structured explanation of why that view was reached

The goal is not to replace PM judgment.
The goal is to improve speed, consistency, and challenge quality in portfolio decisions.

Why this is more than “bull vs bear + judge”

A single agent can sound convincing while missing the best counter-argument.
This framework forces internal disagreement before a conclusion is allowed.

Multi-stage orchestration flow (production logic)

  1. Bull Agent (Haiku): builds the strongest case for a narrative-reality gap.
  2. Bear Agent (Haiku): stress-tests that thesis with timing and counter-evidence.
  3. Expert Agent (Sonnet): adds independent macro context, historical analogs, and reflexivity risk.
  4. Adjudicator (Sonnet): rules on direction, sizing, horizon, and actionable score.
  5. Calibration Agent (Haiku): adjusts raw probability to improve confidence calibration.

This creates practical advantages over single-agent output:

  1. Lower one-sided bias: every thesis is stress-tested before scoring.
  2. Independent expert challenge: both sides are reviewed through a macro-specialist lens.
  3. Better calibration: confidence is corrected after adjudication, not guessed in one pass.
  4. Higher explainability: PMs can inspect the full argument chain and where uncertainty sits.

Why this matters for PMs, CIOs, and risk teams

  1. Faster escalation: identifies events that deserve immediate committee attention.
  2. Cleaner IC process: provides a structured bull-vs-bear brief in one artifact.
  3. Cross-team alignment: discretionary and risk functions review the same rationale.
  4. Portfolio protection: supports earlier hedging and exposure rebalancing.

Evidence snapshot

Metric Result
Historical evaluation set 97 macro-financial events (1971-2024)
Headline accuracy 91.8%
Adjusted accuracy 92.8%
Calibration (Brier) 0.047
Quiet-period FPR 36.8% overall, ~15% on truly calm events

Practical deployment model

Use as a decision-support layer, not autonomous execution:

  1. weekly macro/risk review feed
  2. watchlist prioritization
  3. hedge scenario challenge
  4. pre-trade thesis stress test

PhD research project at RMIT University