Macroeconomist | Monetary Policy & Forecasting
Ph.D. in International Macroeconomics
Paris School of Economics
I am a macroeconomist specializing in monetary policy, forecasting, and macro-financial linkages. My research combines traditional econometric methods with modern machine learning and AI approaches, particularly Large Language Models, to analyze central bank communication and develop improved forecasting tools.
I have research experience at the Paris School of Economics, International Monetary Fund, OECD, and Central Bank of Malta.
Traditional high-frequency identification of monetary policy communication effects operates ex-post, precluding evaluation of alternative strategies before publication. In this paper, we introduce an ex-ante framework that simulates heterogeneous market reactions to central bank communication before release. Our methodology employs Large Language Models (LLMs) to construct an agent-based simulation of 30 synthetic traders with heterogeneous risk preferences, cognitive biases, and interpretive styles. These agents process European Central Bank (ECB) press conference transcripts and forecast Euro interest rate swap rates across 3-month, 2-year, and 10-year maturities. Cross-sectional forecast dispersion provides a model-based measure of market disagreement, validated against realized overnight index swap (OIS) volatility. Analyzing 283 ECB press conferences (June 1998-April 2025), we document Spearman correlations of approximately 0.5 between simulated and realized disagreement, rising to 0.6 under iterative prompt optimization. Results prove robust across prompting strategies, are temporally stable across training and holdout samples, and fare significantly better than simple language complexity scores. For central banks, the framework provides an operational tool to anticipate communication-induced volatility before release, thus enabling ex-ante language refinement. For researchers, it offers a micro-founded alternative to reduced-form event studies, explicitly modeling the heterogeneous interpretive processes through which policy signals are transmitted to asset prices.
This paper investigates the transmission of monetary policy to financial markets within the Euro area, focusing on the role of uncertainty. While previous research has extensively examined the effects of changes in expected policy rates through event studies of European Central Bank (ECB) announcements, the impact of second moments and uncertainty has been far less explored. We address this gap by introducing a novel market-based measure of uncertainty regarding future interest rates, calculated as the difference in the standard deviation of Overnight Index Swap (OIS) rates in a three-day window around ECB policy announcements. Our findings reveal that ECB announcements generally increase market uncertainty about future interest rates, regardless of the sign of the policy surprise. This increased uncertainty significantly impacts asset prices, leading to higher nominal yields, lower stock market returns, and Euro appreciation against safe-haven currencies.
This article explores the use of web-scraped supermarket prices to nowcast food inflation in Malta, where food represents about 16% of the Harmonised Index of Consumer Prices. We compile a novel dataset of over 2,700 products and more than two million daily prices from supermarkets and corner-shops, classified using string-matching and large language models. Three approaches are assessed: a “naïve” benchmark averaging across products, a minimum distance method aligned with official data, and a machine learning framework with mixed-frequency regressions. Out-of-sample tests against the Narrow Inflation Projection Exercise show that web-scraped data can enhance forecast accuracy, but predictive accuracy varies by food categories. Results are preliminary but highlight the value of online prices for real-time monitoring.
This thesis explores different topics related to macroeconomic forecasting. It starts drawing from the literature on international finance some stylised facts on how the global economy became increasingly complex after the demise of the Bretton Woods system. This growing sophistication, in turn, generates a need for more forecasting research to buttress policy-making decisions. Chapter 3 shows that models for financial crises prediction (Early Warning Systems) fare relatively well in terms of forecasting performance. It discusses and compares the role of global and domestic indicators on the materialisation of an external crisis and delves into the difference between missed and predicted crises in terms of relative output losses. Chapter 4 evaluates the performance of short-term forecasts of economic activity produced by the main economic institutions and private sector. It studies how forecast errors correlate to different states of the business cycle, how forecasts produced by different institutions compare to each other and touches on questions related to the political dimension of forecast errors. Finally, Chapter 5 provides a novel dataset of IMF archival documents through a manually compiled dictionary and a term-frequency approach. It measures the depth of discussion about 20 different economic and non-economic crises for the whole Fund membership. It harnesses this rich data source to analyse how the complexity of macroeconomic forecasting has changed over the last decades.
Replication materials for all papers, including data, code, and detailed documentation, are available on my GitHub. I am committed to transparent and reproducible research.