Finding a Needle in a Haystack: A Machine Learning Framework for Anomaly Detection in Payment Systems Staff Working Paper 2024-15 Ajit Desai, Anneke Kosse, Jacob Sharples Our layered machine learning framework can enhance real-time transaction monitoring in high-value payment systems, which are a central piece of a country’s financial infrastructure. When tested on data from Canadian payment systems, it demonstrated potential for accurately identifying anomalous transactions. This framework could help improve cyber and operational resilience of payment systems. Content Type(s): Staff research, Staff working papers Topic(s): Digital currencies and fintech, Financial institutions, Financial services, Financial system regulation and policies, Payment clearing and settlement systems JEL Code(s): C, C4, C45, C5, C55, D, D8, D83, E, E4, E42
Machine learning for economics research: when, what and how Staff Analytical Note 2023-16 Ajit Desai This article reviews selected papers that use machine learning for economics research and policy analysis. Our review highlights when machine learning is used in economics, the commonly preferred models and how those models are used. Content Type(s): Staff research, Staff analytical notes Topic(s): Central bank research, Econometric and statistical methods, Economic models JEL Code(s): A, A1, A10, B, B2, B23, C, C4, C45, C5, C55
Predicting Changes in Canadian Housing Markets with Machine Learning Staff Discussion Paper 2023-21 Johan Brannlund, Helen Lao, Maureen MacIsaac, Jing Yang We apply two machine learning algorithms to forecast monthly growth of house prices and existing homes sales in Canada. Although the algorithms can sometimes outperform a linear model, the improvement in forecast accuracy is not always statistically significant. Content Type(s): Staff research, Staff discussion papers Topic(s): Econometric and statistical methods, Financial markets, Housing JEL Code(s): A, C, C4, C45, C5, C53, D, D2, R, R2, R3
A New Approach to Infer Changes in the Synchronization of Business Cycle Phases Staff Working Paper 2014-38 Danilo Leiva-Leon This paper proposes a Markov-switching framework to endogenously identify the following: (1) regimes where economies synchronously enter recessionary and expansionary phases; and (2) regimes where economies are unsynchronized, essentially following independent business cycles. Content Type(s): Staff research, Staff working papers Topic(s): Business fluctuations and cycles, Econometric and statistical methods, Regional economic developments JEL Code(s): C, C3, C32, C4, C45, E, E3, E32
The Application of Artificial Neural Networks to Exchange Rate Forecasting: The Role of Market Microstructure Variables Staff Working Paper 2000-23 Nikola Gradojevic, Jing Yang Artificial neural networks (ANN) are employed for high-frequency Canada/U.S. dollar exchange rate forecasting. ANN outperform random walk and linear models in a number of recursive out-of- sample forecasts. Content Type(s): Staff research, Staff working papers Topic(s): Exchange rates JEL Code(s): C, C4, C45, F, F3, F31
Forecasting GDP Growth Using Artificial Neural Networks Staff Working Paper 1999-3 Greg Tkacz, Sarah Hu Financial and monetary variables have long been known to contain useful leading information regarding economic activity. In this paper, the authors wish to determine whether the forecasting performance of such variables can be improved using neural network models. The main findings are that, at the 1-quarter forecasting horizon, neural networks yield no significant forecast improvements. […] Content Type(s): Staff research, Staff working papers Topic(s): Econometric and statistical methods, Monetary and financial indicators JEL Code(s): C, C4, C45, E, E3, E37, E4, E44