This is the abstract of an article that is currently in preparation and that I am planning to publish about the use of machine learning in wealth management.


There is a lot of buzz around
artificial intelligence (AI) and machine learning (ML) and their
application to
financial services and specifically to wealth and investment 
management; specific examples include use cases applied to chatbots,
robo advice, and credit risk scoring. Almost daily, articles are
focusing on specific functional areas, but do not address the potential
machine learning from an end to end perspective. They usually do not
provide enough details on which techniques can be applied for which use
case.  This document will attempt  to document use cases by process area
and potential use of ML algorithms
and techniques. For each domain, we will give pratical examples of
techniques that can be used to automate the process or enhance the human
decision making process. Here is a non exhaustive list of areas covered
in the article:

  • Client segmentation and risk profiling
  • Client background checks and onboarding
  • Product On-Boarding and recommendations
  • Advisory process
  • Investment and trading
  • Investment research
  • Client servicing and support
  • Anti money laundering

We will also discuss which of the current techniques can be applied to
which process and give an overview of existing algorithms that can be
used in the investment and wealth management context. The article will
also investigate what is the future of machine learning in wealth
management and what challenges financial services might face while
implementing Machine Learning applications in practice. For example in
light of specific confidentiality needs of wealth management as well as
the availability of data for smaller institutions.



Note: If you want to vote for this article to be published in the AI book, you can register and vote under this link:


Note that I have no commercial interest with this site.

(Visited 4 times, 1 visits today)
%d bloggers like this: