Alternative Data and Artificial Intelligence Techniques

This chapter provides an introductory overview of portfolio management. We introduce (i)the evolution of portfolio management and a typology of portfolio management over the past decade, (ii)the classic asset classes and derivatives in portfolio management, and (iii) traditional and modern approaches for portfolio management. We also introduce common tools for measuring portfolio returns and return variance.

This chapter explores the trends in financial asset management. We summarize the global asset management industry, and discuss current trends across multiple regions, including North America, Europe, and China to illustrate the differences. We also discuss ESG, blockchain, and robo-advisor, which push the market forward as newer technology.

With the progress of science and technology, Machine Learning has made great progress in the financial service industry. Its wide application in all aspects of finance makes it have a great impact on financial markets, financial institutions, and financial supervision. By summarizing the application of Machine Learning in the financial industry and taking Alpha Portfolio research as an example, this paper focuses on the opportunities and challenges faced by the application of Machine Learning in the financial industry.

This chapter explains how individuals, business processes, and sensors produce alternative data. It also provides a framework to navigate and evaluate the proliferating supply of alternative data for investment purposes. It demonstrates the workflow, from acquisition to preprocessing and storage using Python for data obtained through web scraping in order to set the stage for the application of ML. It concludes by providing examples of sources, providers, and applications. After reading this chapter, we hope readers can have a detailed understanding of the application process of alternative data in finance, including the sources of alternative data generation, evaluation criteria, etc. We also provide specific application cases and Python codes for the readers’ reference.

This chapter describes how major companies in different countries are utilizing big data analytics in alternative data domains. There are four regional divisions: the United States, China, Europe, and Asian countries except China. Companies that play a prominent role in alternative data utilization are presented.

As one of the main sources of data, text plays an important role in various fields. This chapter mainly introduces the application of textural analysis in the financial field. Firstly, we introduce two techniques of text analysis, including natural language processing and Machine Learning/Deep Learning. Secondly, we also introduce factors for finance built on textural dataset analysis, which includes readability, tone and sentiment factors, similarity, semantic, uncertainty, accuracy, and popularity. Through this article, we have explained the importance and potential of textural analysis in finance.

Artificial intelligence enables the Internet of Things to acquire perception and recognition capabilities, and the Internet of Things (IoT) provides AI with data for training algorithms. The combination of IoT and AI generates and collects massive data, and stores it in device terminals, edge terminals, or on the cloud. Then, the data can be intelligently analyzed through machine learning, so as to realize the digitalization and intelligent connection of all things. Therefore, in this chapter, we will detail a series of risk measurement models based on IoT and their.

Nowadays, the scarcity of global resources has become extremely prominent, with the demand for sustainable development increasing. Investors have been paying more attention to this perspective, considering sustainably values in their investment decisions and strategies, thus creating responsible investment. Responsible investment, also known as ethical investment, not only advocates financial performance but incorporates company environmental, social, and governance influences. Specific criteria of responsible investment include SRI (Social Responsibility Investing), II (Impact Investing), SI (Sustainable Investing), GF (Green Finance), etc. Soon after, ESG was introduced.

This chapter provides an overview of an important factor in financial markets, sentiment. Market pricing is influenced by how traders react to information and the market and can sometimes be “unfair”. Behavioral finance can explain some “irrational” investment behavior and pricing bias in the market. The sentiment is widely used today, including in the stock and crypto market, and this article briefly introduces sentiment indicators and sentiment analysis methods to evaluate it.

With the trend of increasingly complex big data, how to handle and improve the authenticity of data has become an important issue related to the credibility of data. This chapter discusses how to imitate and detect similar applications and how to identify fake reviews by machine learning and various statistical methods using deceptive applications and fake reviews as examples.

We intend to illustrate profitable trading strategies in the EURUSD exchange rate market using machine learning techniques. Consequently, we applied three supervised learning classification techniques (K-Nearest Neighbors, Support Vector Machines, and Random Forests) in the problem of one day ahead directional prediction of the EURUSD exchange rate with autoregressive terms as inputs. The performance of said machine learning models was benchmarked against two traditional techniques (Naive Strategy and Moving Average Convergence/Divergence). The Random Forest and K-Nearest Neighbors models produced superior results compared to the other models in terms of Net Annualized Returns and Sharpe Ratio.

This chapter provides a real-life case study on how to measure a company’s performance using ESG metrics. Here, we mainly select a specific research object, namely SPAC, and comprehensively describe the whole analysis process. This includes the role of SPAC, the history of its development and supervision, and the external metrics of SPAC.

This chapter investigates the impacts of the ESG factors onto Corporates’ fundamental aspects using healthcare industry as a case study. It first uses the difference-in-difference method to assess COVID-19’s effect on the pharmaceutical industry. The results show that the epidemic-related companies have a significant positive performance in ROA/ROE, compared to the energy and bio-manufacturing companies. It then examines the correlation between the ESG factors and the companies’ economic performance. The fixed effects regression suggests there is a significant negative relationship between total greenhouse emissions and ROA/ROE. Finally, using the generated ESG scores, we constructed an investing strategy in which they buy the ten highest rated companies’ stocks, and short the bottom ten rated companies’ stocks within the portfolio. The outcome shows that for 75% of the time period (excluding the period 2017–2018), the portfolio outperformed the control group in which we bought stocks randomly.

This chapter provides an overview of the relevant content of data visualization, comprehensively introduces its basic knowledge, and common tools to use, and gives a related case for a complete visualization process; this chapter elaborates on the applications of (i) data visualization, (ii) introduction to Python visualization tools, (iii) data distribution chart, and (iv) financial data case analysis.

This chapter studies the interaction between MongoDB database and Python by introducing the background of AWS Lambda. It mainly introduces the knowledge related to the three subjects involved in the topic, and this chapter elaborates the applications of (i) MongoDB and its pricing strategy, (ii) Python, its environment, and modifying codes, and (iii) AWS LAM and its complete usage process. The authors also discuss promising directions of using alternative or unstructured data for both academics and practitioners.