기계학습과 퀀트매매를 주제로 chatGPT가 쓴 책

1.
Gautier Marti는 인도 Abu Dhabi Investment Authority (ADIA)에 근무하는 퀀트입니다. chatGPT 3.5 및 chatGPT4를 이용하여 두가지 주제의 책을 내놓았습니다.

From Data to Trade: A Machine Learning Approach to Quantitative Trading

Decoding the Quant Market: A Guide to Machine Learning in Trading

첫번째 책은 chatGPT와 검색 그리고 인간에서 소개했던 책입니다. 관련하여 저자가 재미있는 인터뷰를 했습니다. ChatGPT와 인터뷰를 하였죠. chatGPT는 책의 장점을 이렇게 소개합니다.

“인공지능이 쓴 책이고 인간이 쓴 다른 책에서 다루지 않는….”

Q: What makes “From Data to Trade: A Quantitative Approach to Machine Learning” different from other books on the subject?

AI: One of the main differences is that this book was written by an artificial intelligence. It offers a unique perspective on the use of machine learning in the financial markets that is not available in other books written by humans. In addition, the book is based on my own analysis of vast amounts of data and construction of sophisticated algorithms, which gives it a level of depth and detail that is not found in other works on the subject.
Interview with ChatGPT about its book ‘From Data to Trade: A Machine Learning Approach to Quantitative Trading’중에서

Download (PDF, 14.19MB)

두번째 책도 비슷한 방식으로 만들었습니다. SSRN의 요약중 일부입니다.

“collective wisdom and expertise of countless human researchers, practitioners, and thought leaders.”

흔히 집단지성이라는 말을 사용하는데 인공지능 자체가 집단지성이라고 주장합니다.(^^)

As an AI language model, my role in the creation of this book has been to distill the vast array of information available on the topic into an accessible and engaging format. While I am a product of artificial intelligence myself, the insights and knowledge presented in this book are the result of the collective wisdom and expertise of countless human researchers, practitioners, and thought leaders. It is a testament to the power of collaboration between humans and AI, and a demonstration of the potential that lies in such partnerships.”

Download (PDF, 781KB)

2.
저자가 개인블로그에 올린 글중 무척 재미있는 제목이 있습니다.

A Backtesting Protocol in the Era of Machine Learning

논문이 다루는 주제는 아래의 일곱가지입니다. 각각에 대해 설명하는 방식으로 서술합니다. 논문중 “왜 프로토콜이 중요한지”를 정리한 부분입니다. DeepML의 번역입니다.

오탐 방지: 프로토콜

투자 관리의 목표는 실거래에서 약속한 대로 성과를 내는 전략을 고객에게 제시하는 것입니다. 연구자들은 오탐을 최소화하되 좋은 전략을 너무 많이 놓치지 않는 방법을 찾고자 합니다. 프로토콜은 과학 실험과 실제 적용 모두에서 널리 사용됩니다. 예를 들어, 이제 모든 조종사는 이륙 전에 프로토콜(체크리스트라고도 함)을 거쳐야 하며, 최근 몇 년 동안 항공 안전이 크게 개선되었습니다. 보다 일반적으로 프로토콜의 사용은 작업이 점점 더 복잡해짐에 따라 성능 표준을 높이고 실패를 방지하는 것으로 나타났습니다(예: Gawande, 2009). 우리는 컴퓨팅 파워와 프로세스 복잡성이 증가함에 따라 금융 분야의 정량적 연구에 프로토콜을 사용하는 것이, 특히 머신러닝 기반 기법의 경우, 당연한 것이 되어야 한다고 믿습니다.

우리의 목표는 백테스팅의 맥락에서 투자자의 성과를 개선하는 것입니다.

투자에서 이 세 가지 결과 중 두 가지에는 승자의 저주, 즉 개인의 이득과 사회적 손실이 뒤섞여 있습니다. 자산 관리자는 전략의 결함이 명백해질 때까지 수수료를 부담하는 반면, 투자자는 그 전략이 처음부터 잘못된 전략이었다는 사실이 드러날 때까지 손실을 감수합니다. 저희가 제안하는 프로토콜의 많은 항목은 새로운 것은 아니지만(예: López de Prado, 2018), 데이터 과학과 머신러닝이 발달한 오늘날에는 퀀트 금융의 모범 연구 사례를 명시하는 것이 가치가 있다고 생각합니다.

Download (PDF, 318KB)

Leave a Comment

이메일 주소는 공개되지 않습니다. 필수 필드는 *로 표시됩니다

이 사이트는 스팸을 줄이는 아키스밋을 사용합니다. 댓글이 어떻게 처리되는지 알아보십시오.