1.
알고리즘 트레이딩과 관련한 일을 합니다. 그렇다고 알고리즘을 연구하지 않습니다. 끊임없이 변화하고 새로운 연구 및 시도들이 나오고 읽지만 적용할 생각을 하지 않습니다. tensor flow로 이런저런 시험을 해보자고 몇 번 작심했지만 tensor flow를 설치하지도 못하고 있습니다. 그럼에도 변화에 뒤지지 않으려고 노력합니다. 쫓아가지 못할지라도…
얼마전 bobsguide에 올라온 Algo trading innovations stoke fear despite increased popularity을 보면 금융산업이 알고리즘트레이딩을 어느 수준에서 도입하는지를 알 수 있습니다. 알고리즘 트레이딩이라는 표현을 사용하였지만 유행하는 단어로 하면 AI입니다. 이중에서 강화학습 알고리즘과 관련한 부분입니다. 저도 두번 정도 소개하였던 내용입니다.
JPMorgan과 Deep Reinforcement Learning
Morgan Stanley와 JP Morgan이 연구하는 RL
Reinforcement learning (RL) algorithms are increasing in popularity
RL algorithms work by penalising algo strategies for making the wrong decision while rewarding them for making the right one. They were championed by JPMorgan in 2018 and play a large role in the future development of algo trading. RL is part of a broader family of AI referred to as deep learning, which fintechs such as Sentient have been developing in recent years.
Algorithms must have a dynamic approach to fluctuating market conditions, and RL algos provide a robust quality to algo trading that has become integral to its success, according to JPMorgan. RL’s impressive self-teaching capabilities take further control from an organisation’s human hands – an impact which has both cynics, wary of too much responsibility being placed on the machine, as well as advocates, who celebrate the lack of emotion with which the systems learn to better enter and exit positions.
이상의 기술은 최근 여의도의 관심을 한 몸에 받고 있는 크래프트테크놀로지스도 이와 관련한 기술을 적용하고 있습니다.
같은 기사중 헤지펀드들이 알고리즘을 어떻게 바라보는지를 조사한 자료를 보면 흥미로운 내용들이 있습니다. 예를 들어 “왜 알고리즘을 도입하느냐?”는 질문에 대한 답변들이 다양합니다. 개인적으로 Improve trade productivity이 가장 많을 줄 예상했는데. Reduce market impact, Execution consistency, Cost 등 여러 요인들이 비슷한 비중을 차지합니다.
좀더 시야를 좁혀서 로보어드바이저와 관련한 부분입니다. Using AI for safekeeping capital investments and portfolio optimisation은 자산운영산업과 관련한 AI영역을 다루고 있지만 그 중에서 프트폴리오 최적화 부분입니다.
Accurate NAV calculations is an essential part of an institution’s fund administration process. AI-based applications automate NAV and pricing calculations, which can also be applied across an institution’s range of funds, including more complex funds like multi-class and master-feeder funds, while also having variation checks and pricing controls.(중략)
When used for portfolio optimisation, AI-based applications seek to minimise risk metrics that use advanced models such as volatility, tracking error, parametrical and historical expected shortfall, Monte Carlo expected shortfall and maximum drawdown. Wealth managers will be able to visualise graphically a portfolio’s components, benchmarks and credit risks, as well as how those credit risks are being reduced.
Reimagining Performance Measurement in an AI World은 자산운용산업내의 Performance Analyst의 새로운 역할을 제시합니다.
That being said, domain expertise and human intuition will be no less valuable in the future. Regardless of how powerful and intuitive AI becomes, domain expertise will be required to outline the objectives of the technology, recalibrate the algorithm when circumstances change, and serve as the arbiter of efficacy. Technological experience will be valuable of course, but if the programmer doesn’t understand the difference between a Sharpe and a Sortino Ratio, they will need to pair themselves with performance veteran who does.
Moreover, the performance analyst of the future will still be counted on to understand the unique needs of clients, and then explain how a specific fund or strategy is positioned to meet these objectives. What are the client’s future funding liabilities? What are their return expectations and risk tolerance for a given allocation, in both the near-term and longer-term time horizons? How does a given strategy complement other fund commitments? And how should they interpret performance at a given point in time, in a given market environment? Natural language generation, while it can transform raw data into a conversational narrative, still lacks the intuition and nuance that characterizes the most successful and valued performance analysts today.
2.
오늘 글을 쓴 이유입니다. The Future of Trading: Technology in 2024 – The Impact of AI, Big Data and Analytics on Your Trading Desk을 소개합니다.
