Ajit Desai - Latest
-
-
Machine learning for economics research: when, what and how
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. -
From LVTS to Lynx: Quantitative Assessment of Payment System Transition
We quantitatively assess the changes in participants’ payment behaviour from modernizing Canada's high-value payments system to Lynx. Our analysis suggests that Lynx's liquidity-saving mechanism encourages liquidity pooling and early payments submission, resulting in improved efficiency for participants but with slightly increased payment delays. -
Improving the Efficiency of Payments Systems Using Quantum Computing
We develop an algorithm and run it on a hybrid quantum annealing solver to find an ordering of payments that reduces the amount of system liquidity necessary without substantially increasing payment delays. -
Macroeconomic Predictions Using Payments Data and Machine Learning
We demonstrate the usefulness of payment systems data and machine learning models for macroeconomic predictions and provide a set of econometric tools to overcome associated challenges. -
Estimating Policy Functions in Payments Systems Using Reinforcement Learning
We demonstrate the ability of reinforcement learning techniques to estimate the best-response functions of banks participating in high-value payments systems—a real-world strategic game of incomplete information. -
Using Payments Data to Nowcast Macroeconomic Variables During the Onset of COVID-19
We use retail payment data in conjunction with machine learning techniques to predict the effects of COVID-19 on the Canadian economy in near-real time. Our model yields a significant increase in macroeconomic prediction accuracy over a linear benchmark model.