Название: Handbook of Artificial Intelligence and Big Data Applications in Investments
Автор: Larry Cao
Издательство: CFA Institute Research Foundation
Год: 2023
Страниц: 164
Язык: английский
Формат: pdf (true)
Размер: 10.1 MB

Artificial Intelligence (AI) and Big Data have their thumbprints all over the modern asset management firm. Like detectives investigating a crime, the practitioner contributors to this book put the latest Data Science techniques under the microscope. And like any good detective story, much of what is unveiled is at the same time surprising and hiding in plain sight.

Each chapter takes you on a well-guided tour of the development and application of specific AI and Big Data techniques and brings you up to the minute on how they are being used by asset managers. Given the diverse backgrounds and affiliations of our authors, this book is the perfect companion to start, refine, or plan the next phase of your Data Science journey.

As a result of recent advances in computing power, data abundance, and cloud services, asset managers are increasingly eager to use Artificial Intelligence (AI) and Machine Learning (ML) solutions to unearth meaningful insights from their data. Like the chicken and egg paradox, the buy side needs advanced technology and established efficient processes to capture and store data in sufficient granularity and requires ML expertise to help sift through the reams of structured and unstructured data. Though buy-side traders still consider sourcing liquidity their biggest challenge—one that statistical algorithms cannot easily solve—they recognize the advantages of ML-assisted approaches. Based on our observations, data delay and capture capabilities are improving and there is enthusiasm for putting existing data to work. ML’s abilities to uncover patterns in large datasets and surface correlations that humans cannot detect have buy-side traders and portfolio managers learning about and turning to open-source ML libraries, such as scikit-learn, tools similar to those used by traditional data scientists. In turn, these budding quasi-data scientists are developing such ML-based solutions as trigger algorithms that can help inform a direction to take when unforeseeable events occur or ML-powered solutions that overcome limitations by using data gathered from other industries.

Contents: