Ai For Alpha Launches Multi Decoding Portfolio, Combining Four Alternative Strategy Engines in a Unified Risk Framework

The new AI-powered portfolio combines Ai For Alpha’s CTA and hedge fund replication strategies with a proprietary Decoding Alpha engine. It invests across 28 liquid markets within a centralized risk framework.

PARIS--(BUSINESS WIRE)--Ai For Alpha, a fintech company specializing in AI-powered investment strategies, today announced the launch of its Multi Decoding Portfolio, a systematic cross-asset strategy designed to combine four complementary alternative investment return engines within a single portfolio-level risk framework.

Multi Decoding is designed to pursue an absolute-return objective by combining differentiated alternative return drivers.

Risk disclosure: The strategy can incur substantial losses, and diversification and portfolio controls do not eliminate investment risk.

Rather than allocating independently to separate strategy sleeves, Multi Decoding looks through to their underlying market exposures, bringing them together within a common investment universe spanning 28 liquid markets across equities, government bonds, currencies, commodities and credit.

This approach is designed to identify overlapping exposures, recognize offsetting positions and manage portfolio risk consistently across the combined strategies.

Multi Decoding expands Ai For Alpha’s proprietary Decoding Suite, a technology platform in live use since 2022 that powers systematic replication strategies across alternative investment exposures including CTAs, systematic global macro, risk parity, private equity and hedge fund benchmarks.

Ai For Alpha licenses its strategies to institutional investors and quantitative investment strategy (QIS) desks at major banks through structures including SMAs and indices.

Four strategy engines, one portfolio

The Multi Decoding Portfolio combines four Ai For Alpha strategies:

• Decoding Alpha — a proprietary strategy derived from Ai For Alpha’s Enhanced Decoding methodology. Enhanced Decoding systematically adjusts allocations relative to the standard replication process, with the objective of improving the portfolio’s risk-return profile.
• CTA Multi-Horizon Decoding — a systematic CTA replication strategy designed to capture both the market exposures and changing trend horizons of a diversified universe of large CTA managers.
• CTA Risk Off — a dynamically estimated CTA portfolio from which long equity exposures are removed, while retaining the ability to hold short equity positions. Its objective is to maintain exposure to systematic CTA return drivers while limiting structural dependence on rising equity markets.
• Global Hedge Funds Decoding — a systematic replication of the aggregate exposures of a diversified portfolio of hedge funds representative of the broader industry.

A unified risk framework

Multi Decoding is differentiated not only by the strategies it combines, but by how they are combined.

Instead of treating each component as an independent allocation, the framework aggregates their underlying exposures across a common market universe. This enables Ai For Alpha to identify when apparently different strategies express similar market views, recognize offsetting positions and manage concentrations at the level of the overall portfolio.

“Combining several alternative strategies does not necessarily create diversification,” said Thomas Jacquot, CFA. “Different strategies can ultimately build exposure to the same underlying market risks. Multi Decoding gives us look-through across the portfolio and allows those exposures to be managed centrally, much closer to the way a multi-strategy manager would manage risk. For institutional investors, the objective is to preserve differentiated return engines while avoiding unintended concentrations.”

Portfolio-level controls include limits on individual market exposures, ranges for aggregate exposures across major asset classes and controls on how quickly positions may change.

The framework is designed to reduce the risk that several underlying strategies unintentionally accumulate the same directional exposure and to preserve diversification across the portfolio’s different return engines.

The portfolio is primarily implemented through liquid market instruments, providing transparency into aggregate positions and the contribution of the underlying strategies.

From replication to Decoding Alpha

A separate innovation within Multi Decoding is Decoding Alpha, a proprietary return engine derived from Ai For Alpha’s Enhanced Decoding framework.

Ai For Alpha’s Standard Decoding process estimates the dynamic market exposures of a reference investment strategy from its historical valuations and represents those exposures through a portfolio of liquid market instruments.

Enhanced Decoding builds on this process by allowing the model to adjust allocations among comparable market exposures, with the objective of improving the portfolio’s risk-return profile.

The difference between the exposures generated by Enhanced Decoding and those generated by the Standard Decoding process forms the Decoding Alpha strategy.

Decoding Alpha therefore seeks to capture a return component generated by Ai For Alpha’s proprietary allocation methodology rather than by replicating the underlying exposures of an external investment strategy.

“Multi Decoding represents an important evolution of our Decoding technology,” said Béatrice Guez, CEO of Ai For Alpha. “We initially developed Decoding to infer the investment decisions embedded in complex strategies. After four years of research and live implementation, we are now using that same expertise to generate a proprietary return engine through Decoding Alpha. This extends Ai For Alpha from replication into proprietary alpha generation and portfolio construction.”

Building on Ai For Alpha’s Decoding research

Multi Decoding builds on Ai For Alpha’s research into systematic investment replication, graphical models, multi-horizon trend strategies and Enhanced Decoding. Its Decoding technology has been implemented through institutional SMAs and bank QIS platforms since 2023.

Multi Decoding extends this approach from the replication of individual strategies to the construction and centralized risk management of a diversified portfolio of alternative investment models.

About Ai For Alpha

Ai For Alpha develops systematic and AI-powered investment models for institutional investors and financial institutions.

Its proprietary Decoding technology uses machine learning to estimate changing market exposures from historical strategy valuations and represent those exposures through systematic portfolios of liquid financial instruments.

Ai For Alpha’s research and investment solutions span CTA, global macro, hedge fund and private equity replication, as well as diversified cross-asset and risk-managed strategies. Its models are designed for institutional implementation through separately managed accounts, indices and bank QIS platforms.

Learn more at aiforalpha.com.

Important Information

This communication is provided for informational purposes only and does not constitute an offer to sell, a solicitation of an offer to buy, or a recommendation regarding any investment product or strategy.

Multi Decoding is a systematic, model-based process. References to “alpha,” “absolute return” and “Risk Off” describe investment objectives, strategy names or methodologies and do not imply or guarantee positive alpha, positive absolute returns, capital protection, outperformance or negative correlation with equity markets.

Model outputs, market exposures and relationships between asset classes may change over time. The portfolio may incur losses, including during periods of declining equity or bond markets. Portfolio controls are intended to manage exposures and risk concentrations but cannot eliminate investment risk or prevent drawdowns.

Investment strategies involving futures, credit instruments and other derivatives involve risk and may result in losses.

Actual implementation depends on factors including market liquidity, portfolio size, turnover, transaction costs and investor-specific constraints.

Published research relating to individual methodologies should not be interpreted as a performance record for the combined Multi Decoding process. This release does not present performance results for Multi Decoding.

There is no assurance that any investment objective will be achieved.