Posts by Collection

portfolio

publications

Modeling the Reserve Demand to Facilitate Central Bank Operations

Published in IMF Working Paper WP/23/179, 2023

A methodology for estimating banks' demand for central bank reserves as a function of the policy rate spread and macro and market-structure variables, comparing parametric sigmoid curves with nonparametric machine-learning estimators, with scenario analysis to guide central bank operations.

Recommended citation: Chen, Z., Kourentzes, N., and Veyrune, R. (2023). "Modeling the Reserve Demand to Facilitate Central Bank Operations." IMF Working Paper No. 2023/179. International Monetary Fund.

Guatemala: Technical Assistance Report—The Statistical Component of Liquidity Forecasting

Published in IMF Technical Assistance Report, 2024

Technical assistance to Banco de Guatemala on forecasting liquidity to calibrate its daily deposit operations, including forecasts for each deposit type in the reserve-requirement base, statistical reconciliation into an aggregate forecast, and a hands-on workshop.

Recommended citation: Gallardo, D., Veyrune, R., and Chen, Z. (2024). "Guatemala: Technical Assistance Report—The Statistical Component of Liquidity Forecasting." International Monetary Fund.

Forecasting Tail Risk via Neural Networks with Asymptotic Expansions

Published in IMF Working Paper WP/24/99, 2024

CoFiE-NN combines an LSTM with Cornish–Fisher expansions to forecast Value-at-Risk. It captures nonlinear dynamics in higher moments while keeping the output interpretable, and outperforms EGARCH-t and Extreme Value Theory models on simulated data and on 22 currencies.

Recommended citation: Sakurai, Y., and Chen, Z. (2024). "Forecasting Tail Risk via Neural Networks with Asymptotic Expansions." IMF Working Paper No. 2024/099. International Monetary Fund.

United States: 2024 Article IV Consultation (Box 7: The Longer Term Size of the Fed’s Balance Sheet)

Published in IMF Country Report No. 24/232, 2024

Box 7 uses the reserve-demand model from WP/23/179 to estimate the reserves needed for the Federal Reserve's ample-reserves regime (US$2.1–3.2 trillion).

Recommended citation: International Monetary Fund (2024). "United States: 2024 Article IV Consultation—Press Release; Staff Report; and Statement by the Executive Director for the United States." IMF Country Report No. 24/232.

Panama: Financial Sector Assessment Program

Published in IMF Country Report No. 24/233 and related FSAP reports, 2024

Contributed to stress testing and modeling of liquidity, solvency, and interconnectedness in the financial sector.

Recommended citation: International Monetary Fund (2024). "Panama: Financial Sector Assessment Program." IMF Country Report No. 24/233.

Liquidity Forecasting—Part II: The Statistical Component

Published in IMF Monetary and Capital Markets Department Technical Assistance Handbook, 2024

A practitioner guide to the statistical side of central bank liquidity forecasting: forecasting autonomous factors (currency in circulation, net foreign assets, government balances) and liquidity demand with ETS and ARIMA models, forecast combination, reconciliation, and forecast evaluation.

Recommended citation: Chen, Z., Kourentzes, N., Lafarguette, R., Panagiotelis, A., and Veyrune, R. (2024). "Liquidity Forecasting—Part II: The Statistical Component." IMF MCM Technical Assistance Handbook. International Monetary Fund.

Liquidity Forecasting and FX Intervention Strategies

Published in IMF Monetary and Capital Markets Department Technical Assistance Handbook (chapters), 2026

Handbook chapters on liquidity forecasting and foreign exchange intervention strategies for central bank practitioners.

Recommended citation: Veyrune, R., Chen, Z., and Lattie, C. (forthcoming). "Liquidity Forecasting" and "FX Intervention Strategies." IMF MCM Technical Assistance Handbook. International Monetary Fund.

talks

teaching

Forecasting Framework for Central Bank Systemic Liquidity

IMF Capacity Development, IMF CD Facilities and Central Banks, 2022

The course teaches advanced statistical methods to forecast systemic liquidity in the context of different operational frameworks. It covers techniques suitable for central banks with fixed exchange rate arrangements, central banks operating an interest rate corridor, and central banks with an interest rate floor system. The IMF has designed a full-fledged software infrastructure for forecasting central banks’ autonomous factors. IMF staff will present the infrastructure during the course. The forecasting framework designed by IMF staff is also tailored for countries with limited data and under-developed financial markets.

Risk-Based Framework for FX Intervention

IMF Capacity Development, IMF CD Facilities and Central Banks, 2023

This five-day course aims to train participants, mostly from central banks, on the theoretical and practical aspects of central bank interventions in foreign exchange markets. Overall, the course aims to provide participants with a solid understanding of the theoretical and practical aspects of central bank interventions in foreign exchange markets, along with hands-on experience in using Python and IMF software infrastructure for modeling conditional foreign exchange intervention triggers. The main takeaways for participants include a deeper understanding of FX interventions, statistical concepts, and Python programming. Participants can use this knowledge to improve their skills in their respective roles and advance their careers in the field of finance.

Tools for Central Banking Implementation: Liquidity Forecasting and FX Intervention

IMF Capacity Development, IMF CD Facilities and Central Banks, 2024

The course teaches advanced statistical methods to estimate and implement foreign exchange intervention rules based on a risk approach, to forecast systemic liquidity in the context of different operational frameworks, and to conduct stress tests on central bank’s balance sheet as well as its profit & loss.

INST314: Statistics for Information Science

Teaching Assistant, College of Information, University of Maryland, 2026

Teaching assistant for Dr. Sarah Lee. I lead R and RStudio labs and discussions, and built an interactive Cookie series of demos on sampling distributions, confidence intervals, and hypothesis testing.