Introducing GeonatiQ

Unlock Alpha in the EU Carbon Market AI-Driven Carbon Price Forecasting Powered by

This presentation is provided for informational purposes only and does not constitute investment advice, an offer, or a solicitation to buy or sell any financial instrument. The carbon market forecasts and scenarios described herein are generated by proprietary AI models and are subject to assumptions, limitations, and uncertainties. Past performance or model outputs are not indicative of future results. Recipients should not rely solely on this material when making investment or business decisions and should undertake their own independent analysis. No representation or warranty, express or implied, is given as to the accuracy, completeness, or reliability of the information contained in this presentation. Neither the authors nor their affiliates accept any liability for any loss arising directly or indirectly from the use of this presentation or its contents.

Executive Summary

Why Predict EU Carbon Prices?

Coming Next – Carbon LLM & Custom Agents • Real-time interpretation of news, regulation, and policy • Scenario testing and adaptive trading strategies • Automated agents: emails, dashboards, reports, execution workflows and a lot more! Who We Serve • Corporates – long-term hedging, compliance, and cost certainty • PE Firms – portfolio-level carbon risk monitoring • Hedge Funds & Traders – quant-driven trading and execution • Banks – emissions finance, structured products, balance sheet exposure • Private Investors – interested to trade EU Carbon Markets

✓ EU ETS = €781bn market – the world’s largest carbon market, yet still underexploited ✓ In 2024, only €11.8bn actively traded – with c.50% of participants inactive due to uncertainty, complexity, or lack of tools ✓ Regulatory tightening + EU ETS II (€705bn) set to drive long-term price rises – Bloomberg projects EUAs will double in price by 2030

Current Service

• Proprietary AI Forecasting Model – predicts EUA price trends 1–9 months ahead. • 6-Month Model : +12.0% return | Sharpe 1.77 | 71% win rate • 3-Month Model : +8.1% return | Sharpe 4.35 | 100% win rate

Why Now? 10 Reasons the Market is Ripe

2 1

1

6

1

Largest global carbon market – 1.1 Gt CO₂ coverage

€781B market cap, 11.8B allowances traded

4 3

3

2

7

+28% YoY volume growth despite price dip

Structural tightening begins in 2026

6 5

5

8

3

No new EUA volume until Sept 2025 = regulatory clarity

BloombergNEF forecast: €149/t by 2030 (vs €66.5/t in 2024)

1 4

7

8

9

232 MtCO₂e abatement by 2030

€705B in climate revenues from ETS II (2027–2035)

3 5

9

10

1 10

c.50% of firms who hold EUAs don't trade = untapped market

AI = the gateway for activating new participants

Who Is This For? Unlocking Edge Across the Market

Audience

Use Case

Hedge Funds

Tactical alpha, weekly forecasts, options overlays

Banks

Structured products, ESG-linked solutions

Commodity Traders

Improved futures timing and pricing

Corporates

Long-term hedge optimisation

Asset Managers

ESG fund allocation insight

Hot Off the Press Our Breakthrough in Commodity Price Forecasting • Published academic journal paper on AI meta-learning for carbon price prediction with Imperial College London • Paper No. 1 ranked in Commodities & Energy Economics • Real-world case: EU ETS • Outperforms traditional models using signal decomposition, deep learning, and meta-learning • Empirical validation of AI's superiority in long-horizon trading

The 50% Opportunity Half the Market Doesn't Trade. That’s an ‘In’ Market Participation Landscape

• c.50% of EUA holders do not actively trade due to timing uncertainty, lack of knowledge, risk aversion • ~3,515 companies covered under the EU ETS • ~10,000 installations (e.g. power plants, factories, airports). One company can operate multiple installations

GeonatIQ Advantage

• Full mapping of EU ETS emitters, direct outreach network available • AI-powered price prediction closes the confidence gap • Drives participation and unlocks dormant alpha

What We've Built

Carbon Price Prediction Engine Powered by Meta-Learning

Trading Layer •

Built-in directional signals

• Strategic ‘pause’ triggers during low-confidence windows Results: Reduced drawdowns, capital preservation, false-signal avoidance

• Inputs : Futures, macro data, energy prices, regulations, COT • Architecture : Multi-scale signal decomposition + adaptive training • Outputs : High-accuracy 1–9 month EUA price predictions

Prediction Summary Model initially Trained and Tested on Dec ‘24 Futures Contract Data

Performance – Dec24 Contract: 10.3% 9M (Biweekly)

Tools Applied:

12.0% 6M (Weekly)

•

MA overlays

•

Signal overlays

•

Options simulations

Insights:

Sharpe 1.53 | Win Rate: 66.7%

Sharpe 1.77 | Win Rate: 70.8%

•

Calibration boosts returns

8.1% 3M (Monthly)

1.8% 1M (Weekly)

•

Shorter windows = higher frequency success

•

Sharpe >2 = v. strong performance

Sharpe 4.35 | Win Rate: 100%

Sharpe 1.67 | Win Rate: 75.0%

Prediction Summary Accurate EUA price forecasts across multiple timeframes, using 14-day moving averages of actual vs predicted prices:

1-Month Forecast: • MAE = €0.90 • High precision for tactical short-term trades 3-Month Forecast: • MAE = €2.40 • Optimised for swing trading and quarterly planning 6-Month Forecast: • MAE = €3.00 • Supports medium-term positioning and risk management 9-Month Forecast: • MAE = €2.93 • Captures long-term price direction with low deviation MAE (Mean Absolute Error) shows the average difference between predicted and actual prices – lower is better. Note: Additional datasets added since 2024 training have improved MAE accuracy by ~20% per forecast window.

Why December Contracts? A: It's the Market Anchor

• Aligns with annual

compliance cycles (April) • Dominates open interest (c.70-80%) – market anchor • Deepest liquidity, tightest bid ask spreads • Primary benchmark for pricing, trading, and risk models • Preferred by funds, corporates, and compliance buyers

Open Interest Breakdown:

Dec 2025: 328,566 (72%)

Aug 2025: 47,796 (10%)

June 2025: 36,685 (8%)

Dec 2025

Aug 2025

June 2025

Introducing Our Carbon LLM Next-Gen Market Interpretation Layer

Our Carbon-specific LLM interprets news-driven market response by understanding context , not just sentiment.

Interpret policy, energy, and macro context

Simulate "what if" scenarios

Model causal price impacts beyond sentiment

Integrate with numerical model for hybrid prediction

Deliver foresight for traders, corporates, regulators and others

Examples:

" EU delays CBAM" → impact simulation

•

• "Coal subsidy in Poland" → projected EUA shift

Carbon LLM cont…

How it works: Decompose Carbon Prices & Identify News Attention Our LLM breaks down carbon prices into high frequency, low frequency, and long-term residual components. It then learns which type of news each component attends to the most, extracting signals that traditional models miss.

Insights from R&D:

High-frequency signals Correspond to macro/market sentiment headlines (e.g., bear market fears).

Low-frequency signals Relate to structural/partnership news (e.g., Amazon rainforest monitoring projects).

Residual/long-term signals Map to monetary policy & central bank updates (e.g., ECB rate decisions).

So?

A 5-month news attention history was used to predict 30-day forward prices. Achieved MAE: 2.2 EUR, with strong directional accuracy.

Key takeaway:

The LLM “self-learns” how different news types map to market horizons, creating a powerful, explainable forecasting engine for EU ETS carbon prices.

Our AI Agents for Automation

Private Equity Firms Examples: ➢ Consolidated view of carbon exposure across all portfolio companies ➢ Automated scenario analysis on regulatory changes and carbon price impact ➢ Alerts when rising carbon costs threaten EBITDA or exit valuations Banks Examples: ➢ Carbon-linked indices, swaps, and structured products powered by automated feeds ➢ Real-time monitoring of liquidity and regulatory updates for trading desks ➢ Automated exposure mapping across client flow and balance sheet positions

Corporates Examples: ➢ Automated hedging signals to lock in allowance costs at the right time ➢ Live dashboards showing compliance spend vs. budget in real time ➢ Instant alerts when allowance prices breach internal cost thresholds Hedge Funds & Traders Examples: ➢ Momentum, volatility, and spread signals delivered directly to traders ➢ Automated workflows that connect signals to trade execution systems ➢ Real-time dashboards for P&L and risk, with automated stop loss alerts

Every build is bespoke – combining reporting, automation, and execution in ways that fit each client’s strategy and pace.

AS

Engagement Options From Intelligence to Actionable Insight

Various engagement options to fit your needs:

1

2

3

Structured Research Report • Tailored to your timeframe, goals, and exposure • Add custom training data • Consider custom scenarios

Bespoke GUI / Agent Build • Custom GUI built for each client for data selection and queries • Our agent platform automates workflows - from on-demand reports and live dashboards to market commentary, with LLMs advising on the trading and hedging tactics

Performance Access •

For funds/banks/corporates

•

Access model for minimal cost in exchange for GeonatIQ P&L exposure Plus (through out sister firm, AIP ) : connect with LPs & GPs active in emissions-focused strategies seeking to invest in ESG opportunities and platforms

•

Summary Massive Market. Minimal Tools. • Yet c.50% of allowance holders don't trade – timing & complexity are key barriers • Analysts project EUA prices could double by 2030, presenting a rare alpha opportunity • EU ETS = €781B market, €11.8B allowances traded (2024)

For Who?

• Hedge Funds, Banks, Traders, Corporates, Private Investors

• Use cases: Alpha generation to long-term carbon hedging

Our Solution: The

AI Forecasting

Stack

• Multi-scale signal decomposition for short-, mid-, and long-term forecasts

• LLM designed to integrate news context into price prediction and simulate scenarios

• Agents for workflow automation, on-demand reporting, and strategy

et in Touch

AI-powered foresight for commodities price prediction

“This isn't just a model. It's an edge”

Email clients@geonatiq.com

Website www.geonatiq.com

Key Contact Mike Schuil (mike@geonatiq.com)

Appendix: AI Driven Meta-Learning for Long Term Commodity Price Forecasting: A Case Study of European Markets (EU ETS)