Markets by Trading view

Integrating More Data, Analytics, And Systemic Thinking Into A Fundamental Investment Process


Hosting over 200 people the annual Quant Strats Europe event is one for the diary if you are a quant or a portfolio manager. Focusing on the buy-side the event attracts delegates and speakers from across the world and not just London. The editorial team from Disruption Banking were just one of the media partners who came to find out more about the biggest trends to affect asset managers and hedge funds today. Fundamental investment analysis and process formed big parts of the agenda.

We decided to focus on a panel moderated by Igor Yelnik, CIO and CEO of Alphidence Capital. Igor was joined on stage by 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.

To kick off proceedings, Igor challenged the panellists about the history of data and investment. Michael responded by sharing how many of the things that portfolio managers are doing today haven’t changed since the 1950s.

Hitendra added how in the last four to five years he has seen a change in sentiment where risk has moved to the front office. He also alluded to factor investing to minimize exposure to market swings. Factors can impact the performance of a portfolio because they hinder the risk management process.

Answering Igor’s next question about when to use data as a portfolio manager, Hitendra answered by explaining how less jargon and more emphasis on the question and how to interpret the solution to that question was needed.

Large Language Models and Fundamental Investing

It was later in the panel that the conversation turned to all things artificial intelligence and Large Language Models (LLMs). The audience also sat up in interest.

Igor raised the question about whether LLMs have changed the way that portfolio managers use data. Ihsan responded by highlighting three of the things he feels are relevant today.

  1. Search is a typical component of a typical workflow, and it is getting easier with the help of LLMs.
  2. Because LLMs are good at memorizing information, they can take existing models made by brokers specific to a given industry sector. This can substantially shorten the time it takes to develop a good model.
  3. The challenge is where data is presented in a different format by different institutions and brokers. Some data is inaccessible due to the way that information is published by firms. Some firms don’t want the data to be widely available as there is a cost to the production of the data.

Ultimately, when brokers initially created the templates for the information that they share today they didn’t think about what machines would need to be able to read that data. Everyone’s format is different. Graphs and models themselves are accessible for LLMs, Igor highlighted, but when it comes to text-based data this still holds a challenge for how data is being analysed and used in a fundamental investment process.

About Alphidence

Igor Yelnik launched Alphidence in 2020.

The strategy behind Alphidence was created in 2012 – 2013 and traded live at ADG from 2013 to 2019. Prior to ADG Igor was the Head of Portfolio Management and Research at IPM Informed Portfolio Management where he developed and managed a substantially similar strategy. The AUM of the strategy both at IPM and ADG was in the $ billions.

Alphidence Capital Limited is authorised and regulated by the FCA. The firm is CFTC/NFA registered as a Commodity Pool Operator (CPO) and Commodity Trade Advisor (CTA).

Alphidence Capital manages a Cayman domiciled Alphidence Systematic Macro Fund and can take multiple managed accounts.

The Investment Strategy

The investment strategy is systematic. It uses a number of return forecast factors routed in the economic and financial theory. These return forecast factors, alongside with the risk management techniques utilised by the strategy are key to the performance of the strategy.

The fund objective is to deliver gross of fees excess return of around 13-15% with a Sharpe ratio of around 0.9-1 over the long term. The fund also aims to exhibit low correlations with traditional asset classes and other hedge fund strategies, over the long-term, including on the tails of the distribution of returns.

The strategy takes both directional and relative value positions in futures and FX.

Return forecasts are grounded in the economic and finance theory and systematically calculated based on economic and market data. The return forecasts are produced in five independent multi-factor models. Four of these models are relative value and one – directional. Each factor targets a specific premium or market anomaly. The factors are intuitive and expected to deliver positive returns over time.

A risk allocation process combines the factors and models into the overall portfolio taking into account systematically implemented risk constraints. The risk management part of the strategy aims at giving priority to the factors and models which are expected to deliver higher returns and at reducing the tails of the distribution of return.

The trading universe consists of 42 listed futures and 9 currencies (implemented via 8 FX Forwards). The fund typically holds positions in all these instruments.

Long Term Strategy

The strategy is long-term and has a low turnover. It runs daily and makes marginal changes to portfolio positions. The ability to capture a ‘patience premium’ leads to using less crowded and more persistent factors, which means that the more tenacious inefficiencies are used. This also means that the strategy has a higher capacity.

The Fund trades markets globally. It takes positions only in developed countries where highly liquid instruments are available. The fund has neither long nor short bias.

It is the job of the research team to constantly look for possible ways to further improve the strategy. The team endeavours to make such changes in a thoughtful and gradual manner thus ensuring the stability of the strategy and avoidance of style drift.

Author: Andy Samu


The Editorial Team at #DisruptionBanking have taken all precautions to ensure that no persons or organisations have been adversely affected or offered any sort of financial advice in this Article.

This Article is most definitely not Financial Advice.

Our readers are reminded that investing in cryptocurrencies can mean that you will lose all your money.

See Also:

How Quantitative Strategies Can Help Hedge Funds | Disruption Banking

‘Our Portfolios Have So Many Moving Parts’: GAM’s Farida Mustafazade on the Challenges of a Quant | Disruption Banking

How Reinforcement Learning can be Revolutionary for AI-driven Alpha Strategies | Disruption Banking

The Unconventional Way of Generating Alpha | Disruption Banking

Leave a Reply

Your email address will not be published. Required fields are marked *

Related Posts


Write your email to verify subscription


Sign up for our free newsletter and receive the latest banking and fintech stories, straight to your inbox - every week