Our lab applies proven methodologies to new datasets and helps to contextualize the results within the realities of a contemporary Canadian perspective, an aspect poorly covered by current literature.
For the lab’s empirical work, we use a spectrum of datasets that encompass millions of data points for hundreds of thousands of working Canadians. Through the data, we develop data-driven analytic techniques that allow us to identify complex relationships in variables from large datasets – starting with attributes associated with their employment, gender, age, etc. and moving downstream to deductions from income (taxes, saving rates, benefits, etc.).
We use advanced data analytics techniques which will allow us to identify non-linear, complex interactions. More specifically (or technically) we will conduct a suite of advanced clustering (k-means, k-prototypes and PAM) and t-stochastic neighbour embedding visualization (principal components, t-SNE) techniques suited to specific data types.
Financial Wellness Lab: State of the Nation
The purpose of this paper is to create a framework to be used by the stakeholders associated with the Financial Wellness Lab for engagement and collaboration. The framework seeks to document what we know so far, what we don't know, and where we plan to go.
Learning About Financial Health in Canada
This paper applies cluster analysis to eleven (continuous) years' worth of responses to the Canadian Payroll Association (CPA) survey of employed Canadians. The clustering algorithm clearly identifies three distinct groups of respondents. Between-group comparison of response patterns reveals that two of the groups lie on opposite sides of the financial health spectrum, and leads us to label their members "financially stressed" and "financially capable", respectively.
Measuring the Gap Between Elicited and Revealed Risk for Investors: An Empirical Study
This paper proposes a novel methodology for comparing an individual's elicited and revealed risk. We propose using Value-at-Risk to measure elicited and revealed risk and the discrepancy between them, showing whether clients are over-risked or under-risked. We demonstrate the methodology using a dataset from a Canadian private financial dealer.
Know Your Clients’ Behaviours: A Cluster Analysis of Financial Transactions
We show that the KYC information—such as gender, residence region, and marital status—does not explain client behaviours, whereas eight variables for trade and transaction frequency and volume are most informative. Hence, our results should encourage financial regulators and advisors to use more advanced metrics to better understand and predict investor behaviours.
The Impact of Saving on Financial Resilience: Keeping It Simple
In this paper, we take a deeper dive into the issue of saving. To do so, we examined unique dataset(s) of investor transactions to determine the relationship between investor behaviours, household savings, and investment outcomes. We examined real-world observed behaviours through advanced data analytics in the form of machine learning to explore previously unknown patterns (referred to as clusters) and seek a determination of any causal relationships.
Enhancing an existing algorithm for small-cardinality constrained portfolio optimization
Recently, a new algorithm was developed to find CCEFs for small cardinalities. Relative to other algorithms for this problem, this algorithm is very intuitive, and its authors demonstrated that it performs at nearly the state-of-the-art. However, we have found that the algorithm seems to struggle in certain situations, particularly when faced with both bonds and equities.
Kernel Metric Learning for Clustering Mixed-Type Data
Distance-based clustering and classification are widely used in various fields to group mixed numeric and categorical data. Our approach improves clustering accuracy when utilized for existing distance-based clustering algorithms on simulated and real-world datasets containing pure continuous, categorical, and mixed-type data.
An Overview of Machine Learning for Asset Management
Machine learning has been widely used in the asset management industry to improve operations and make data-driven decisions. This article provides an overview of machine learning for asset management by presenting various machine learning models in the context of their applications, and highlights the challenges of implementing machine learning in asset management.
The Complexity of Financial Wellness: Examining Survey Patterns via Kernel Metric Learning and Clustering of mixed-type Data
Recent market events and inflation have significantly affected the financial stress facing many individuals, but understanding the main stressors is paramount to supporting them in making better long-term financial decisions. Financial advisors must understand the types of stress their clients face to provide tailored advice.