The KYOS commodity VaR engine provides clear insight into potential future losses, enabling risk managers to take timely and informed action. Based on the results, users can adjust portfolio positions, execute new hedging transactions, or allocate additional capital as a risk buffer.
Calculate risk on a portfolio of commodity contracts and assets
Choose between analytical, Monte Carlo simulation and historical methods
Backtest of VaR with historical P&L movements
Show VaR on sublevels per book, commodity or counterparty
Present marginal VaR, component VaR, price & volatility sensitivities
Foresee potential losses
Value-at-Risk (VaR) is a widely used measure to quantify market risk in a portfolio of positions. It estimates the potential change in the market value of a portfolio over a given time horizon at a specified confidence level. For example, a 10-day 95% VaR of USD 1 million indicates that the portfolio is not expected to lose more than USD 1 million over a 10-day period in 95% of the cases.
The KYOS VaR engine is built specifically to capture commodity transactions, including futures, forwards, options, and structured contracts with price indexations and time spreads.
Within the KYOS platform, VaR can be monitored alongside market risk and actual P&L developments over time. This enables users to track whether observed P&L movements are consistent with the modelled risk levels. With a 95% confidence level, losses exceeding the VaR threshold are statistically expected about once every 20 days. Backtesting functionality allows users to verify whether historical P&L outcomes align with this expectation.
Understand VaR drivers
The main drivers of Value-at-Risk are portfolio positions and the price volatility of the underlying commodity markets. The KYOS VaR model provides detailed insight into both factors by displaying positions and volatility across time periods, helping users understand how risk evolves within the portfolio. VaR results can be presented in flexible formats, allowing users to identify the main sources of risk by book, commodity, or time period.
To further analyze the drivers of portfolio risk, the model provides several detailed outputs. These include Marginal VaR, which shows the impact of an additional trade on total portfolio VaR; VaR by sub-selection, such as book, counterparty, or commodity; and VaR by cross-selection, providing a breakdown across combinations of books and commodities. Component VaR and sensitivities on prices and volatilities enable users to quickly identify where the largest risk contributions originate.
Breach notifications
Breach notifications are an essential feature for risk managers. Each day, the model compares the calculated VaR with the relevant VaR limits. Limits can be defined at the total portfolio level as well as for sub-levels such as individual books or commodities.
If a VaR limit is exceeded, the system immediately generates a clear alert. Users are notified through email and through the platform’s user interface, enabling timely action to manage and reduce risk exposure.
Methodologies
KYOS supports three widely used approaches to calculate Value-at-Risk: parametric VaR, historical simulation, and Monte Carlo simulation. This allows users to select the methodology that best fits their portfolio characteristics, risk policies, and regulatory requirements.
For Monte Carlo-based VaR, KYOS offers two advanced simulation approaches. Users can apply PCA-based simulations, which efficiently capture correlations between market variables, or use KySim, the proprietary KYOS price simulation model designed for realistic energy market price dynamics.
KyVaR summary
Calculate risk on a portfolio of contracts and assets
Choose and compare VaR with method of your preference: parametric, Monte Carlo simulation and historical simulation
Breakdown of VaR in various cross-selections and periods to identify the risk drivers
No unexpected surprises: receive VaR-breaches in automated workflows
Marginal VaR shows the impact of an extra trade on company risk
Better informed decisions
KYOS software
KYOS helps companies to make better decisions in complex commodity markets by combining deep sector expertise with robust technology. Our solutions deliver actionable insights and measurable value that can be used in day-to-day operations.
Advanced Analytics That Drive Better Decisions
Turn complex data into clear insights that consistently lead to smarter, faster and more informed decisions.
Stable and Reliable by Design
A robust, proven platform you can depend on — even in demanding, high-stakes environments.
Seamless Workflow Integration
Integrates smoothly into your existing processes, enabling automation and improving operational efficiency.
Powerful and Flexible API
A comprehensive API library allows effortless connectivity with your internal systems, tools, and data sources.
Intuitive and Easy to Use
A clean, user-friendly interface designed for professionals — no unnecessary complexity, no steep learning curve.
Outstanding Support and Measurable Results
Expert support that understands your business and helps you translate analytics into tangible performance improvements.
Non-parametric Forward Looking Value-at-Risk
The paper ‘Non-parametric forward looking Value-at-Risk’ is written by Marcus Nossman and Anders Vilhelmsson, and published in Journal of Risk, June 2014.
This paper proposes a new model for computing Value-at-Risk forecasts. The model is fully non-parametric and easy to implement. Further, it incorporates information about the markets perceived uncertainty about the future. The forward looking information is obtained from the option market via CBOEs implied volatility index VIX. Using SP500 data from 1990-2010 we fond that the use of option implied volatility compares favorably to GARCH type models in terms of forecast performance. By comparing the model primarily used in the banking sector to our new model, we find that a financial institution using our model has on average a lower market induced capital requirement (MCR). However, during the time period leading up to the financial crises our model gives a 40% higher MCR.
Download directly this article
A Forward-Looking Approach to Value-at-Risk
Traditional Value-at-Risk (VaR) models have been widely criticised for relying heavily on historical data and restrictive assumptions, which can limit their ability to capture future market risks. This paper introduces a new non-parametric and forward-looking VaR model (HS‑VIX) that incorporates market expectations of volatility derived from option prices, specifically the VIX index.
The key objective is to improve risk measurement by combining historical simulation with forward-looking information, avoiding assumptions about return distributions while still capturing changing market conditions.
Methodology: Combining Historical Data with Implied Volatility
The proposed model builds on historical simulation techniques but adjusts past returns using implied volatility from the VIX index. This allows the model to reflect current market expectations of future uncertainty, rather than assuming that past volatility patterns will persist unchanged.
Unlike traditional models such as GARCH, the HS‑VIX approach is fully non-parametric and easier to implement. It also mitigates issues related to volatility clustering by dynamically scaling historical returns according to current volatility levels.
Performance and Empirical Results
Using S&P 500 data from 1990 to 2010, the study compares the HS‑VIX model with standard historical simulation and GARCH-based approaches. The results show that the HS‑VIX model performs better overall, particularly in capturing risk dynamics and avoiding clusters of extreme losses.
Importantly, the model leads to lower average capital requirements over the full sample, but significantly higher requirements (around 40% more) in the period leading up to the financial crisis. This suggests that the model is more sensitive to rising market risks and could provide earlier warning signals.
Implications for Risk Management
The findings highlight the value of incorporating forward-looking information into risk models. Financial institutions using the HS‑VIX framework would likely hold more capital during periods of increasing uncertainty, potentially improving resilience during crises.
Overall, the paper demonstrates that a non-parametric, forward-looking VaR model can outperform traditional approaches by better aligning risk estimates with market expectations. This has important implications for both regulators and financial institutions seeking more robust risk management tools.
Commodity exposure hedging Interview
Treasury Today: KYOS’ Director for Corporate Clients, Richard Cornielje, spoke with Tom Alford of Treasury Today. This interview centers on understanding the importance and difficulties of commodity exposure hedging. This UK based firm publishes a monthly magazine as well as online content about all aspects of treasury management.
Download directly this article
Introduction: Rethinking Commodity Risk Management
This interview with Richard Cornielje, as reported by Treasury Today, emphasizes a fundamental shift in how companies should approach commodity risk management. Rather than attempting to predict future prices, Cornielje argues that firms should focus on preparing for uncertainty by understanding their full exposure. While hedging is widely used, traditional approaches often treat each commodity in isolation, ignoring the interconnected nature of modern production processes. He suggests that companies can significantly improve decision-making by adopting a multi-commodity perspective on risk.
The Problem with Simplistic Approaches
A key issue identified in the interview is the oversimplification of commodity budgeting and risk assessment. Many companies rely on basic methods, such as adding a fixed percentage to last year’s prices, which fails to reflect the high volatility of commodity markets. Cornielje highlights that commodity price volatility can be several times greater than foreign exchange volatility, meaning that even modest changes can lead to substantial cash flow swings. This is particularly relevant for industries such as food and beverage production, where multiple commodities, from energy to packaging materials, contribute to overall costs.
Understanding Multi-Commodity Exposure
Cornielje stresses the importance of analyzing how different commodities interact within a product. For example, when choosing between sucrose and dextrose as inputs, a company might initially prefer the cheaper option. However, deeper analysis reveals that these inputs are tied to different underlying markets, e.g. sugar and wheat, which have different volatility profiles. A slightly more expensive input may lead to lower overall risk and more stable cash flows. This demonstrates that procurement decisions should consider not only price but also the risk characteristics of underlying commodities.
Implications for Hedging Strategy
A more comprehensive understanding of commodity exposure leads directly to improved hedging strategies. Cornielje explains that some commodities offer limited hedging opportunities due to short-term market liquidity, while others, supported by deeper markets, can be hedged over longer horizons. For instance, materials like aluminum may appear more expensive initially but provide greater flexibility in long-term hedging compared to alternatives such as tinplate. Effective hedging therefore requires aligning procurement choices with market liquidity and risk tolerance.
Role of Data and Advanced Analytics
The interview highlights the growing importance of integrated data systems and advanced analytics in managing commodity risk. By consolidating information across contracts, markets, and products, companies can better assess correlations between commodities and simulate price scenarios. Techniques such as Monte Carlo simulation enable firms to translate commodity price movements into value-at-risk and cash-flow-at-risk metrics. This allows hedging strategies to be based on robust data rather than assumptions, improving both transparency and effectiveness.
Conclusion
The interview concludes that effective commodity hedging requires moving beyond simplistic, single-commodity approaches toward a holistic, data-driven framework. By understanding the relationships between commodities, incorporating market volatility, and leveraging advanced analytics, companies can design more resilient hedging strategies and better manage financial risk.
Enterprise Risk Management in Utilities
Energy Risk features an article on ERM (Enterprise Risk Management), focusing on the importance of such a tool for utilities, October 2016.
For many utilities in North America and Europe, Risk Management (ERM) has long been little more than a regulatory compliance exercise. Yet as markets become more volatile and business models more uncertain, ERM is becoming critical to utilities’ success. Maybe even survival. Unfortunately, the fact remains that in most companies, ERM functions remain neglected. They are still largely confined to their original internal compliance role.
Download directly this article
Introduction: The rise of Enterprise Risk Management
This article explores the growing importance of enterprise risk management (ERM) among utilities facing declining revenues, rising costs, and structural changes in energy markets. ERM aims to provide a holistic view of risk across the entire organization, helping firms identify long-term trends and strategic threats such as the rise of renewables and distributed generation. By integrating different sources of risk, utilities hope to make better decisions and capture value in a challenging business environment.
Challenges of implementing ERM
Despite its potential, implementing ERM effectively remains difficult. Many utilities still manage risks in isolated “silos,” with separate approaches for trading, operations, and strategic planning. Cyriel de Jong notes that ERM is still underdeveloped in most energy firms, with many organizations relying on basic risk registers rather than fully integrated frameworks. The challenge lies in combining quantitative models with broader strategic insights, as focusing too much on either can lead to incomplete risk assessments.
From fragmented to integrated risk management
A central issue highlighted in the article is the lack of aggregated risk metrics across organizations. Different business units often use inconsistent assumptions, time horizons, and methodologies, making it difficult to assess total exposure. De Jong emphasizes the need for enterprise-wide measures such as earnings-at-risk or cash-flow-at-risk that combine market, volume, and liquidity risks. Without such integration, companies struggle to understand how external factors, like mild winters or price shocks, affect overall performance and cannot take appropriate action.
Limits of quantitative models
The article also discusses the limitations of relying solely on quantitative risk models. While these models are effective for market and trading risks, they become less useful when addressing broader operational or strategic risks. Risk managers must recognize when to complement analytics with qualitative tools such as scenario analysis and expert judgment. Historical examples, including major industry disruptions, show that focusing too narrowly on models can lead to missed opportunities or catastrophic decisions.
Balancing short-term and long-term risks
Utilities face the ongoing challenge of balancing short-term risk management with long-term strategic analysis. According to de Jong, firms often prioritize day-to-day risk due to resource constraints, even though the most significant risks typically emerge over longer horizons. Incorporating long-term perspectives, supported by both data and judgment, can provide valuable insights into enterprise-wide risks and inform better hedging and investment decisions.
Organizational and cultural factors
Finally, the success of ERM depends heavily on organizational alignment and culture. Senior management support is essential to ensure that risk considerations are embedded in strategic decisions. ERM must be seen as a value-adding function rather than a compliance exercise. Clear communication, consistent metrics, and demonstrated success stories help build trust and encourage collaboration across business units.
Conclusion
The article concludes that ERM is an evolving discipline that combines quantitative analysis with strategic thinking. While challenges remain, utilities that successfully integrate risk across their organizations can improve decision-making and better navigate the uncertainties of modern energy markets.
The use of ERM has to move on – do not use it for regulatory compliance only, experts say. Cyriel de Jong: “With a few exceptions, ERM is not well developed in most energy firms. Many companies would now like to make improvements to their existing functions”.
KYOS is your trusted partner in energy and commodity management. We specialize in a range of software solutions for the valuation and optimization of natural gas storages, power plants and renewable assets, such as solar- and wind farms and battery storages. In addition we offer a C/ETRM system for companies that have a large exposure to a variety of commodities, and who want to manage their risks.