Fraud prediction using machine learning: The case of investment advisors in Canada
Bibliographic record
Abstract
The paper contributes to a growing body of empirical work on regulatory technology by proposing machine learning models to detect fraud in financial markets. The recent spate of investment fraud in Canada has exposed regulators’ inability to protect vulnerable investors and the financial markets from financial abuse. As evident by the numerous regulatory task force commissioned in the past two years, Canadian regulators have been looking for ways to detect and prevent fraudulent activities before they occur and support enhanced enforcement powers. The purpose of this study is to use data collected from the Investment Industry Regulatory Organization of Canada (IIROC) to build a machine-learning algorithm to predict fraud in the Canadian securities industry. Data for this project were collected from IIROC’s tribunal cases covering June 2008 to December 2019. In total, 406 cases were retrieved from the IIROC’s website. The results from four machine learning models reveal that across all the features, the amount of money invested and whether the offender was from a bank-owned investment firm were the high predictors of fraud in terms of the standardized coefficient. Branch managers and regulators should pay careful attention to portfolios that continuously incur losses as a sign of potential fraud. The findings are particularly relevant to regulators seeking new and effective fraud detection techniques while providing enhanced clarity to Canada’s financial markets’ self-regulation.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.019 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.003 | 0.005 |
| Science and technology studies | 0.004 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".