Skip to content

kevinmday/IntentionalTradingSystem

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

141 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

MarketMind — Narrative Propagation Engine

MarketMind is a narrative-driven signal engine that detects emerging intention fields in information systems and models how those signals propagate through financial markets.

Traditional trading systems treat price as the primary signal.

MarketMind reverses that model.

Instead of reacting to price movement, the system monitors narrative signals (news, filings, research announcements, macro events), extracts referenced assets, models propagation across related symbols, and converts those signals into quantified intention scores.

The working assumption is simple:

Narrative propagates before capital moves.

MarketMind attempts to detect that propagation before it fully manifests in price.

Financial markets serve as a real-time proving environment for a broader mathematical framework describing intention propagation in complex systems.

Core Idea

Markets move when aligned intentions propagate through networks of participants.

MarketMind models this process as:

Narrative → Intention Signal → Propagation → Capital Flow → Price

Price movement is treated as a downstream manifestation of narrative propagation, not the primary signal.

System Architecture

MarketMind operates as a continuous event-processing pipeline.

Narrative Sources

Feed Ingestion

Feed Aggregator

Symbol Extraction

Propagation Engine

Diffusion Graph Engine

PsiQuanta Scoring

Runtime Executor

Trade Coordinator

FastAPI API Layer

React / Vite UI

Key Capabilities

Narrative ingestion from multiple sources

Symbol extraction from narrative events

Propagation modeling across related assets

Diffusion graph learning between symbols

Intention-based signal scoring (PsiQuanta)

Continuous runtime signal evaluation

Real-time monitoring dashboard

Repository Structure

IntentionalTradingSystem/

marketmind_engine/

 api/ – FastAPI server

 narrative/ – Narrative ingestion & parsing

 intelligence/ – Propagation / diffusion modeling

 quant/ – Quantitative scoring engines

 runtime/ – Runtime orchestration

 execution/ – Trade coordination & execution

 data/ – Data providers (price feeds)

regime_ui/ – React / Vite dashboard

docs/ – Theory and engineering documents

README.md

Core Engine Components

Narrative Engine

Collects signals from narrative sources including:

news feeds

research announcements

corporate disclosures

regulatory filings

macro events

Narratives are normalized into a common internal event structure.

Symbol Extraction

Narrative events are scanned for references to:

equities

sectors

technologies

macro assets

Each reference becomes a projection event associated with a symbol.

Propagation Engine

Tracks narrative signals over time and models how influence spreads across assets.

Signals maintain state and decay unless reinforced by new narrative events.

Key metrics include:

FILS (Future Intention Likelihood Scale)

UCIP influence factors

drift metrics

decay functions

Diffusion Graph

MarketMind automatically learns relationships between assets by observing narrative sequences.

Example inferred relationships:

NVDA → SMCI

AI infrastructure → semiconductor suppliers

Energy policy → oil majors

Defense spending → aerospace contractors

Graph edges strengthen when repeated narrative patterns occur.

PsiQuanta Scoring

PsiQuanta converts narrative propagation signals into quantitative scores.

Inputs include:

narrative signal strength

propagation patterns

market drift

price confirmation

Output:

psi_score

Used by the runtime executor to evaluate potential trades.

Runtime Executor

The runtime system coordinates the entire pipeline.

Responsibilities:

run narrative ingestion cycles

update propagation state

compute intention scores

generate candidate trade signals

Trade Coordinator

Evaluates candidate signals and applies trading rules.

Trade logic focuses on identifying emerging intention fields rather than purely technical price patterns.

Backend Server

The backend runs as a FastAPI application.

Start server:

uvicorn marketmind_engine.api.server:app --reload --port 8001

Example API endpoints:

/api/rss_events

/api/propagation_snapshot

/api/engine/status

/api/engine/last

Frontend UI

The UI is built with:

React

Vite

TypeScript

Tailwind

Location:

regime_ui/

Start UI:

cd regime_ui

npm install

npm run dev

Open in browser:

http://localhost:5173

Development Workflow

Typical development cycle:

git status

git add .

git commit -m "message"

git push

Recommended environment:

Python 3.12

Node.js

FastAPI

React / Vite

Continuous Narrative Monitoring

Narrative signals propagate regardless of market hours.

MarketMind therefore runs in two operational modes:

Narrative Engine → 24/7

Trading Engine → market hours

Narrative state is maintained continuously so emerging signals are captured early.

Research Context

MarketMind is both:

a trading system

a research platform for studying intention propagation

Stock trading provides a high-frequency measurable environment where narrative signals can be evaluated against real outcomes.

The architecture is intentionally context-agnostic and may apply to other domains where signals propagate through networks.

Potential applications include:

financial markets

technology adoption

geopolitical dynamics

social network contagion

information warfare modeling

Project Status

Active experimental system.

Current engineering focus:

narrative propagation modeling

diffusion graph learning

expanded narrative source aggregation

runtime observability

scoring model refinement

Author

Kevin Mark Day

MarketMind is part of a broader exploration into intention-field dynamics in complex systems.

License - Glass Box Open Architecture

Experimental research software

License TBD

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Packages

 
 
 

Contributors