Optimisation of client trust by evolutionary learning of financial planning strategies in an agent based model

In this paper we demonstrate how evolutionary learning can be employed to learn a fuzzy knowledge base (FKB) utilised in strategic decision making among financial planning advisors (FPAs). This is accomplished by building an agent based model (ABM) simulating the process of wealth management between clients and financial planning advisors. The agent based model is implemented in Java using RePast (RePast URL) as the agent based modelling toolkit. Due to the complex nature of the decision making process, fuzzy logic (FL) is used to represent behavioural rules of the agents in this multi-agent based simulation. We show how the proposed evolutionary learning process could be employed in maximising the clients' trust towards their FPA as well as their own financial wellbeing. Towards the end of the paper, the results obtained using the RePast based simulation are presented showing the advantages of the evolutionary learning process in the context of optimising FPA-client relationships and how it helps to adapt FPA strategies to reduce the number of clients, who leave their FPA and invest elsewhere