The New AI-powered replication strategy is designed to improve benchmark tracking and risk-adjusted performance
Ai For Alpha, a fintech company specializing in AI-powered investment strategies, announced the launch of CTA Multi-Horizon Decoding, a new futures-based replication strategy designed to track the SG CTA trend benchmark with greater precision, transparency, and implementation efficiency.
Read More on Fintech : Global Fintech Interview with Baran Ozkan, co-founder & CEO of Flagright
CTA Multi-Horizon Decoding introduces Ai For Alpha’s new Double Decoding framework, built to capture both asset allocation and changing trend-horizon exposure in CTA benchmarks.
The strategy expands Ai For Alpha’s proprietary Decoding Suite, a technology platform live since 2022 that powers transparent, cost-efficient replication portfolios across alternative investment exposures, including CTAs, systematic global macro, risk parity, private equity, and hedge fund benchmarks. Ai For Alpha licenses these strategies to institutional investors and quantitative investment strategy (QIS) desks at major banks.
CTA Multi-Horizon Decoding builds on the firm’s earlier CTA Decoding strategy, launched in 2022, which delivered one of the strongest benchmark correlations in its category. The new model introduces a two-stage Double Decoding framework designed to replicate not only the benchmark’s asset allocation, but also its changing allocation across trend horizons.
Ai For Alpha’s research found that the time-horizon mix embedded in CTA benchmarks evolves over time as managers adjust to changing market conditions. More recently, the firm observed reduced allocations to medium term windows. CTA Multi-Horizon Decoding was developed to capture these shifts more effectively and improve long-term replication accuracy.
“CTA managers do not keep a static allocation to trend horizons over time,” said Thomas Jacquot, Chief Revenue Officer. “More recently, we have observed stronger allocations to short- and long-term windows. Our earlier model was not explicitly designed to detect those changes. With the multi-Horizon Decoding, we can now capture them more effectively. In our tests, the new model improves replication accuracy and strengthens long-term performance.”
“Allocators are looking for CTA exposure that is transparent, investable, and adaptable across market environments,” said Béatrice Guez, CEO of Ai For Alpha. “CTA Multi-Horizon Decoding builds on our replication expertise while improving the strategy’s information ratio and giving clients greater visibility into the performance drivers.”
A two-stage replication framework
CTA Multi-Horizon Decoding uses a two-step process designed to capture both market trends and changes in manager behavior over time.
In Stage 1, Five Ai for Alpha decoders run in parallel, with each decoder calibrated to a distinct trend horizon spanning from 20 to 500-day time windows. These decoders allocate across 24 liquid futures-based proxies representing key CTA exposures, including equities, rates, foreign exchange, and commodities. Each decoder builds a portfolio tailored to its assigned time window, allowing the model to isolate and capture trend behavior across multiple market horizons.
In Stage 2, the five resulting trend sleeves are dynamically reallocated through a second decoding process. This additional layer is designed to infer the benchmark’s changing mix across time horizons and adjust the strategy, accordingly, producing a more adaptive and representative replication of the current CTA opportunity set.
Institutional implementation
CTA Multi-Horizon Decoding is implemented exclusively through liquid listed futures, offering daily transparency, scalability, and cost efficiency for institutional investors. Ai For Alpha licenses its replication portfolios to institutional allocators and bank QIS platforms for implementation through separately managed accounts (SMAs) or indices.
Clients can monitor portfolio composition and performance over time, including the contribution of each trend horizon to the overall allocation. The strategy is built using a consistent framework for transaction costs, roll methodology, and management fees, and does not rely on look-ahead information. All allocations are based solely on information available at the time decisions are made, supporting a replication process that is both robust and transparent.
Catch more Fintech Insights : Real-Time Payments and the Redefinition Of Global Liquidity
[To share your insights with us, please write to psen@itechseries.com ]
