Back to Library

Improving Charting Decision Making for Stock Market Investors Using Collaborative Agents

by Amal Khaseeb

Quick summary -- an in-depth PhD-level extended summary (10-30 pages) for this book is coming soon.

Improving Charting Decision Making for Stock Market Investors Using Collaborative Agents

By Amal Khaseeb

Quick Summary

A master's thesis from Birzeit University (2008) that proposes a multi-agent system architecture for improving stock market charting decisions. The thesis combines fundamental analysis, technical analysis, and multiple stock-picking strategies (including CAN SLIM, GARP, value investing, and income investing) into a collaborative agent framework that automates the evaluation and recommendation process for investors.

Executive Summary

This work is an academic thesis submitted for a Master of Scientific Computing degree at Birzeit University, supervised by Prof. Ghassan Al-Qaimari and Dr. Naser Abdelkarim. It addresses the challenge of integrating multiple analytical approaches -- fundamental analysis and technical analysis -- into a unified decision-support system for stock market investors. The author proposes a collaborative multi-agent system where specialized software agents each perform distinct analytical functions (fundamental screening, technical charting, strategy evaluation) and then aggregate their findings to produce investment recommendations. The thesis covers the theoretical foundations of both fundamental and technical analysis, reviews multiple investment strategies, and presents the system design, implementation, and testing.

Core Thesis

Individual investors face information overload and cognitive limitations when trying to combine fundamental and technical analysis. A system of collaborative software agents -- each specializing in a particular analytical domain -- can improve the quality and consistency of investment decisions by automating data gathering, analysis, and recommendation synthesis. The multi-agent architecture allows for modular, extensible design where new strategies and analytical methods can be incorporated.

Chapter-by-Chapter Analysis

Chapter 1: Introduction

Establishes the problem: stock market decision-making requires integration of multiple analytical approaches, which overwhelms individual investors. Proposes the multi-agent system as a solution.

Chapter 2: Share Trading

Provides a comprehensive overview of fundamental vs. technical analysis:

  • Fundamental Analysis: Covers quantitative analysis (financial ratios including P/E, P/B, ROE, debt-to-equity) and qualitative analysis (management quality, business model, competitive advantages).
  • Technical Analysis: Covers chart types, trend identification, support and resistance, moving averages, and technical indicators.
  • Investment Strategies: Details five major strategies -- value investing, growth investing, GARP (Growth at a Reasonable Price), income investing, and CAN SLIM (William O'Neil's seven-factor methodology).

Chapter 3: Collaborative Agents

Reviews multi-agent system theory, agent communication protocols, and collaborative decision-making frameworks. Establishes the theoretical basis for using software agents in financial analysis.

Chapter 4: System Design

Presents the architecture of the proposed multi-agent system:

  • Data Collection Agent: Gathers financial data from various sources.
  • Fundamental Analysis Agent: Evaluates companies using financial ratios and fundamental criteria.
  • Technical Analysis Agent: Performs charting analysis using indicators and pattern recognition.
  • Strategy Agents: Implement specific investment strategies (CAN SLIM, value, growth, GARP, income).
  • Coordination Agent: Aggregates recommendations from all agents and produces unified investment decisions.

Chapter 5: Implementation and Testing

Describes the technical implementation and provides test results demonstrating the system's ability to screen stocks and generate recommendations.

Key Concepts and Frameworks

  1. Multi-Agent System (MAS) -- Software architecture where autonomous agents with specialized functions collaborate to solve complex problems.
  2. Fundamental vs. Technical Analysis Integration -- The perennial challenge of combining bottom-up financial analysis with price/volume pattern analysis.
  3. CAN SLIM Methodology -- William O'Neil's seven-factor stock selection system (Current earnings, Annual earnings, New products/management, Supply and demand, Leader or laggard, Institutional sponsorship, Market direction).
  4. GARP (Growth at a Reasonable Price) -- A hybrid strategy combining growth and value investing principles.
  5. Agent Collaboration Protocols -- Communication and decision aggregation methods for multi-agent systems.

Practical Applications for Traders

  1. Automated Screening -- The multi-agent approach demonstrates how traders can systematically screen stocks across multiple criteria without manual effort.
  2. Strategy Comparison -- Running multiple strategies in parallel helps identify stocks that score well across different analytical frameworks.
  3. Bias Reduction -- Automated agents reduce the cognitive biases that individual investors bring to their analysis.
  4. Extensible Framework -- New analytical methods and strategies can be added as additional agents.

Critical Assessment

Strengths

  • Provides a thorough review of fundamental and technical analysis methods
  • The multi-agent architecture is a sound approach to handling analytical complexity
  • Covers multiple investment strategies in a comparative framework
  • Demonstrates practical implementation

Limitations

  • As a master's thesis, the scope is limited and the testing is not extensive
  • The system was designed for a specific time period and market conditions
  • Real-world implementation would require robust data feeds and continuous updating
  • Limited backtesting or out-of-sample validation
  • Academic rather than practitioner-oriented; lacks the depth of specialized trading books

Context

This is an academic thesis rather than a traditional trading book. Its value lies primarily in its systematic comparison of investment strategies and its demonstration of how technology can integrate multiple analytical approaches.

Key Quotes

  • "Individual investors face information overload and cognitive limitations when trying to combine fundamental and technical analysis."
  • "CAN SLIM defines seven factors and implements a combination between value, growth, fundamental and technical analysis all together."
  • "GARPers are similar to growth investors in concentrating on the growth and growth potential of the companies but are more conservative when it comes to taking risk."

Conclusion

While this thesis lacks the depth and market experience of dedicated trading books, it provides a useful academic perspective on the challenge of integrating multiple analytical frameworks for investment decision-making. Its primary contribution is the demonstration that multi-agent systems can effectively combine fundamental analysis, technical analysis, and multiple investment strategies into a coherent decision-support tool. For traders, the most valuable takeaway is the systematic framework for evaluating stocks across multiple dimensions rather than relying on a single analytical approach.