AI in Finance: Challenges, Techniques and Opportunities
, Vol. 1, No. 1, Article . Publication date: June 2021.
centers, over service counters, physical or online interviews and questionnaires about a company,
product or service. (12) Simulation data: collected from simulations about the functionalities,
behaviors and performance of a market, product or service, e.g., the data collected from an artificial
cryptocurrency simulation system or the testbed of a new product listing. (13) Third-party data:
collected by third parties about an underlying product, service, institution or participants, e.g., the
Bloomberg event-driven feeds, or data about relevant third-party products, services, institutions,
or participants.
3.2
Economic-Financial Data Challenges
The above EcoFin businesses and data are coupled with each other in reality. This poses various
opportunities and challenges for data-driven AIDS research [30, 31] in finance and economics. Here,
we categorize them into the following perspectives that synergize EcoFin businesses and their
data. (1) Innovation challenges: e.g., AIDS techniques for inventing novel, efficient, intelligent and
sustainable mechanisms, products, services and platforms. (2) Business complexities: such as AIDS
techniques for representing, learning and managing the intricate working mechanisms, structures,
interactions, relations, hierarchy, scale, dynamics, anomaly, uncertainty, emergence and exceptions
associated with a market, a product or participants. (3) Organizational and operational complexities:
such as AIDS techniques to characterize and improve the diversity and personalized services of
individual customers and sectoral demands, the departmental and institutional coherence and
consensus in operations and services, and the inconsistent and volatile efficiency and performance
in organization and operations. (4) Human and social complexities: such as AIDS techniques for
modeling and managing the diversity and inconsistency of participants’ cognitive, emotional and
technical capabilities and performance and for enabling effective communications, cooperation
and collaboration within a department and between stakeholders. (5) Environmental complexities:
such as AIDS techniques for modeling and managing the interactions with contextual and envi-
ronmental factors and systems and their influence on a target business system and its problems.
(6)
Regional and global challenges: such as understanding and managing the relations between an
economy entity and its financial systems with the related regional and global counterparts and
stakeholders and their influence on the target problems. (7) Data complexities: such as extracting,
representing, analyzing and managing data quality issues, misinformation and complicated data
characteristics, e.g., uncertainty, extremely high dimensionality, sparsity, skewness, asymmetry,
and heterogeneity and couplings (i.e. non-IIDness) [27, 41]. (8) Dynamic complexities: such as
modeling, predicting and managing evolving but nonstationary behaviors, events and activities of
individual and batch markets, products, services and participants. (9) Integrative complexities: e.g.,
systematically modeling and managing the various aspects of the above complexities that are often
tightly and loosely coupled with each other in an underlying EcoFin system.
In conclusion, the EcoFin businesses, data and their challenges discussed in Sections 2 and 3
pose numerous opportunities to the AIDS communities and smart FinTech. Below, we focus on
reviewing the related techniques for data-driven AIDS research in finance and economics. This
review complements the one in [33] that mainly takes a business application perspective.
4
AN OVERVIEW OF AI RESEARCH IN FINANCE
The AIDS techniques to support the aforementioned EcoFin businesses and process their data
are very comprehensive, diversified and evolving. Such techniques address various aspects of
business needs and problems, as reviewed in [33]. Here, we categorize the main AIDS techniques
for smart FinTech into the following groups and briefly summarize their relevant work. Fig. 2 shows
the overall classification of AIDS in finance. (a) Mathematical and statistical modeling: including
numerical methods, time-series and signal analysis, statistical learning, and specifically random