Key Takeaways
- • Quant Strats Europe, an annual event hosting over 200 attendees focused on the buy-side, featured a panel moderated by Igor Yelnik, CIO and CEO of Alphidence Capital, on integrating data, analytics and systemic thinking into fundamental investing.
- • Panelists Ihsan Saracgil (Visible Alpha), Michael Schewitz (Investec Bank) and Hitendra D Varsani (MSCI) said risk management has moved to the front office over the past four to five years, with factor investing used to limit exposure to market swings.
- • Ihsan Saracgil identified three LLM-driven gains: faster search within workflows, faster model development by adapting broker sector models, and persistent difficulty with inconsistent data formats and restricted data access.
- • Alphidence Capital's systematic macro strategy targets gross excess returns of around 13-15% with a Sharpe ratio of roughly 0.9-1 over the long term, using five multi-factor models trading 42 futures and 9 currencies.
- • The article is an editorial piece from Disruption Banking, not a fund disclosure, and reports no new fund launch, AUM change or allocation.
Summary
At the annual Quant Strats Europe event, which drew more than 200 attendees and delegates from around the world, Disruption Banking reported on a panel moderated by Igor Yelnik, CIO and CEO of Alphidence Capital. Panelists were Ihsan Saracgil, Principal Data Scientist at Visible Alpha; Michael Schewitz, Co-Portfolio Manager, Credit Trading and Investment at Investec Bank; and Hitendra D Varsani, Managing Director at MSCI. The discussion covered the history of data in investing. Schewitz noted that many portfolio manager practices have not changed since the 1950s. Varsani said that over the last four to five years sentiment has shifted, with risk moving to the front office, and he pointed to factor investing as a way to limit exposure to market swings.
On large language models, Saracgil identified three points: search is a routine part of analyst workflows and is becoming easier with LLMs; LLMs can absorb broker sector models, which can substantially shorten model development time; and data remains hard to use because institutions and brokers publish it in different formats, and some firms restrict access because producing the data is costly. Yelnik observed that LLMs handle graphs and models well, but text-based data remains a challenge for fundamental analysis.
The article also profiles Alphidence Capital, which Yelnik launched in 2020. Its strategy was developed in 2012-2013 and traded live at ADG from 2013 to 2019; Yelnik previously ran a substantially similar strategy at IPM Informed Portfolio Management, where AUM was in the billions of dollars. The firm is authorised and regulated by the FCA and registered with the CFTC/NFA as a Commodity Pool Operator and Commodity Trading Advisor. It manages a Cayman-domiciled Alphidence Systematic Macro Fund and can run managed accounts. According to the article, the fund targets gross-of-fees excess returns of around 13-15% with a Sharpe ratio of about 0.9-1 over the long term, and aims for low correlation with traditional assets and other hedge fund strategies, including in the tails of the return distribution. Five multi-factor models, four relative value and one directional, generate return forecasts, and a risk allocation process combines them under systematic constraints. The universe consists of 42 listed futures and 9 currencies, implemented via 8 FX forwards. The strategy is low-turnover, takes positions only in developed markets, and has no net long or short bias.
On large language models, Saracgil identified three points: search is a routine part of analyst workflows and is becoming easier with LLMs; LLMs can absorb broker sector models, which can substantially shorten model development time; and data remains hard to use because institutions and brokers publish it in different formats, and some firms restrict access because producing the data is costly. Yelnik observed that LLMs handle graphs and models well, but text-based data remains a challenge for fundamental analysis.
The article also profiles Alphidence Capital, which Yelnik launched in 2020. Its strategy was developed in 2012-2013 and traded live at ADG from 2013 to 2019; Yelnik previously ran a substantially similar strategy at IPM Informed Portfolio Management, where AUM was in the billions of dollars. The firm is authorised and regulated by the FCA and registered with the CFTC/NFA as a Commodity Pool Operator and Commodity Trading Advisor. It manages a Cayman-domiciled Alphidence Systematic Macro Fund and can run managed accounts. According to the article, the fund targets gross-of-fees excess returns of around 13-15% with a Sharpe ratio of about 0.9-1 over the long term, and aims for low correlation with traditional assets and other hedge fund strategies, including in the tails of the return distribution. Five multi-factor models, four relative value and one directional, generate return forecasts, and a risk allocation process combines them under systematic constraints. The universe consists of 42 listed futures and 9 currencies, implemented via 8 FX forwards. The strategy is low-turnover, takes positions only in developed markets, and has no net long or short bias.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.