Nicolas Woloszko (Class of 2010, Humanities and Social Sciences) Honored with the Chancellerie of the Universities of Paris Prize

We are very pleased to share this excellent news: Nicolas Woloszko, alumnus of the École normale supérieure from the Department of Humanities and Social Sciences (class of 2010), has been awarded the Prix solennel de thèse of the Chancellerie of the Universities of Paris at the 2025 edition.

He received the fourth prize in economic sciences for his dissertation entitled “Essays on Nowcasting: Machine Learning, High-Frequency Data and Economic Forecasting”, defended in January 2024 at the ThEMA laboratory of CY Cergy Paris University.

His work lies at the intersection of economics, machine learning, and real-time data analysis. It notably led to the development of the OECD Weekly Tracker, an innovative tool that makes it possible to monitor the weekly evolution of GDP for 48 countries. This tool is now used by numerous international institutions, including central banks, and is widely cited in the economic press.

In his academic path, Nicolas emphasizes the importance of cross-disciplinary perspectives, recalling how his education at ENS Paris-Saclay and at ENSAE nurtured his ability to move between economics, mathematics, and computer science.

A researcher at Capital Fund Management, he now continues his work in a multidisciplinary research environment, in collaboration with physicists, mathematicians, and computer scientists, remaining faithful to an approach that breaks down disciplinary boundaries in economic science.

This distinction recognizes a path of excellence and once again illustrates the ability of normaliens to produce high-impact research closely connected to contemporary issues.

Our warmest congratulations to Nicolas Woloszko for this prestigious recognition.

Summary of his dissertation:

COVID-19 abruptly accelerated the need for public policy actors to have access to real-time data on economic activity. This event hastened the culmination of earlier research on the use of new methods in economics, centered on statistical learning and big data. While the massive datasets created and held by firms contain real-time information about the economy, their processing requires a specific approach based on non-linear learning algorithms.

The present research introduces the OECD Weekly Tracker, a tool for monitoring economic activity that provides real-time estimates of weekly GDP for 48 countries. It relies on Google Trends data, which describe changes in the topics of interest of users of Google Search, a search engine. The main methodological innovation lies in the implementation of a non-linear panel model. A neural network jointly models the relationship between GDP and Google Trends data for 48 countries, while allowing for the existence of disparities across countries in this relationship.

The Weekly Tracker has been updated every week since the summer of 2020. Consequently, the analysis of its historical performance shows that its forecasts were more reliable than those of the Economic Outlook, the OECD’s flagship publication, during the pandemic years. By studying the numerous press articles citing the Weekly Tracker figures, it is also shown that this tool provided indications that were qualitatively accurate and relevant for guiding public decision-making.

The policy relevance of the Weekly Tracker lies both in its real-time publication and in the production of weekly GDP estimates. The series available since 2004 make it possible to enrich retrospective policy analysis by exploiting high-frequency statistical identification methods. The Tracker is used in an article published in the journal Nature Communications, which studies the consequences of the introduction of the health pass in France, Italy, and Germany. Among other results, it reveals that vaccination effort was positively correlated with economic growth, and that the health pass made it possible to gain 6 billion euros of GDP in France, and respectively 1.4 and 2.1 billion euros in Germany and Italy.