🔒
Sign In
Creates investment portfolios around emerging themes (Artificial Intelligence, Electric Vehicles, Clean Energy, etc.) by identifying companies whose business descriptions contain relevant keywords. Equal-weights positions for diversification.
This strategy builds portfolios around investment themes using keyword matching: 1. Theme Definition: Each theme has associated keywords (e.g., AI: 'artificial intelligence', 'machine learning', 'neural network', 'deep learning') 2. Company Scanning: Searches company descriptions, SEC filings, and business classifications for …
Best for: Investors who want exposure to emerging trends without picking individual winners. Works well as satellite positions around a core portfolio. Best for those with conviction …
Techniques Used
Keyword tagging
Equal-weight portfolios
Sign in to access this strategy
Sign In
🔒
Sign In
Uses AI-powered text embeddings to understand semantic meaning and find companies truly related to investment themes, even if they don't use exact keywords. Weights positions by relevance strength rather than equal-weighting.
This strategy improves on keyword matching by understanding semantic meaning: 1. Theme Embedding: Convert theme description into a vector representation (e.g., 'Electric vehicles revolutionizing transportation') 2. Company Embedding: Convert each company's business description into a vector 3. Similarity Calculation: Compute …
Best for: Investors who want sophisticated thematic exposure beyond simple keyword matching. Ideal for those comfortable with AI/ML concepts. Works well for nuanced themes that are hard …
Techniques Used
NLP embeddings
Exposure weighting
Sign in to access this strategy
Sign In
🔒
Sign In
Builds a network graph of company relationships (suppliers, customers, partners, competitors) to find all companies connected to a theme across the entire value chain. Uses Large Language Models to classify companies and detect theme lifecycle stage (emerging, growing, mature, declining).
This advanced strategy maps entire thematic ecosystems using graph analysis: 1. Seed Identification: Start with core theme companies (e.g., Tesla, Rivian for EV) 2. Relationship Extraction: LLM analyzes news and filings to find: - Suppliers (who provides components/materials) - Customers …
Best for: Institutional investors and quantitative funds with infrastructure to build and maintain knowledge graphs. Ideal for capturing entire value chains around major themes. Best suited for …
Techniques Used
Knowledge graphs
LLM classification
Network analysis
Sign in to access this strategy
Sign In