Skip to content

Repository files navigation

Quantitative Deep-Tech Investment Screening Framework

A quantitative investment decision framework developed to evaluate and rank leading deep-tech companies for a hypothetical $1,000,000 venture capital investment.

The project combines multi-criteria decision analysis (MCDA) techniques with quantitative validation to identify the strongest long-term investment opportunity across 49 companies spanning Artificial Intelligence, Robotics, Defence Technology, Space Technology, and Quantum Computing.


Project Overview

Selecting investments in the deep-tech sector is challenging because companies differ significantly in:

  • Business maturity
  • Financial transparency
  • Commercialization stage
  • Technology readiness
  • Risk profile

Traditional financial metrics alone are insufficient to compare private and public companies within the same investment universe.

To address this challenge, this project develops a Quantitative Deep-Tech Investment Screening Framework that integrates financial, technological, strategic, and qualitative indicators into a unified investment evaluation methodology.


Objectives

  • Evaluate 49 leading deep-tech companies
  • Compare companies using 31 evaluation criteria
  • Rank investments using multiple decision-making techniques
  • Validate recommendation robustness using simulation
  • Recommend the strongest candidate for a hypothetical $1 million investment

Methodology

The investment framework follows the pipeline below:

49 Deep-Tech Companies
        │
        ▼
Data Collection
        │
        ▼
Feature Engineering
        │
        ▼
31 Evaluation Criteria
        │
        ▼
Data Normalization
        │
        ▼
Weighted Scoring
        │
        ▼
Analytic Hierarchy Process (AHP)
        │
        ▼
TOPSIS Ranking
        │
        ▼
Monte Carlo Simulation
        │
        ▼
Sensitivity Analysis
        │
        ▼
Investment Committee Review
        │
        ▼
Final Investment Recommendation

Evaluation Criteria

The framework evaluates companies across multiple investment dimensions, including:

  • Financial Strength
  • Technology Leadership
  • Commercialization
  • Execution Capability
  • Competitive Positioning
  • Scalability
  • Investment Risk

A total of 31 quantitative and qualitative features were used to generate the final rankings.


Decision-Making Techniques

This project combines multiple quantitative methods:

  • Weighted Scoring Model
  • Analytic Hierarchy Process (AHP)
  • Technique for Order Preference by Similarity to Ideal Solution (TOPSIS)
  • Monte Carlo Simulation
  • Sensitivity Analysis

Using multiple approaches reduces dependence on a single ranking methodology and improves confidence in the final recommendation.


Key Results

Investment Universe

  • 49 Deep-Tech Companies
  • 5 Strategic Sectors
  • 31 Evaluation Criteria

Top Ranked Investment

Rank Company Weighted Score TOPSIS
1 SpaceX 85.85 0.703
2 Anduril Industries 84.91 0.691
3 OpenAI 84.36 0.684
4 Anthropic 83.92 0.678
5 Figure AI 83.51 0.671

Robustness Validation

The recommendation was validated using Monte Carlo simulation.

Metric Result
Probability of Rank #1 64.9%
Probability of Top 3 91.5%
Probability of Top 5 96.5%

Sensitivity analysis further showed that SpaceX remained Rank 1 across all weighting strategies, indicating that the recommendation is robust to reasonable changes in evaluation assumptions.


Repository Structure

DeepTech-Investment-Framework/
│
├── deeptech_investment_framework.ipynb
├── Investment_Memorandum.pdf
├── README.md
├── requirements.txt
│
└── outputs/

Technologies Used

  • Python
  • Pandas
  • NumPy
  • Matplotlib
  • Plotly
  • SciPy
  • Scikit-learn
  • StatsModels
  • MissingNo
  • Yahoo Finance

Installation

Clone the repository:

git clone https://github.com/DebadattaLiku/DeepTech-Investment-Framework.git

Install dependencies:

pip install -r requirements.txt

Launch Jupyter Notebook:

jupyter notebook

Open:

deeptech_investment_framework.ipynb

Report

The complete investment memorandum is available in:

Investment_Memorandum.pdf

Future Improvements

Potential extensions include:

  • Automated real-time market data integration
  • ESG scoring
  • Reinforcement Learning for portfolio optimization
  • Bayesian weight estimation
  • Explainable AI for investment recommendations
  • Interactive Streamlit dashboard

Author

Debadatta Panda

M.Tech – Industrial Mathematics and Scientific Computing

Indian Institute of Technology Madras

GitHub: https://github.com/DebadattaLiku


About

Quantitative investment screening framework for evaluating 49 deep-tech companies using MCDA, AHP, TOPSIS, Monte Carlo simulation, and sensitivity analysis.

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages