Natural language processing (NLP) is undergoing a revolution as big data and large-language models transform the capacity to represent and analyse textual information and extra signals and meaning.
This conference aims to bring together recent research using these approaches in economics.
14h - Opening remarks
14h05 - Opening speech -“Future Challenges for Text-as-Data in Economics”
Professor Stephen Hansen, University College London
Session 1 – Using NLP to improve forecasting and understand narratives
14h40 - Making text count: Economic forecasting using newspaper text (Kalamara, Turrell, Redl, George, Kapadia, 2022)
15h00 - Risky news and credit market sentiment (Labonne and Thorsrud, BI Norwegian Business School)
15h20 - The impact of monetary surprises on exchange rates: insights from a textual analysis approach on a panel of countries (Bricongne and Marolleau, Banque de France)
15h40 – Mining the Gap: Extracting Firms’ Inflation Expectations From Earnings Calls, (Albrizio, Dizioli and Vitale Simon, 2023, IMF)
This paper constructs a new cross-country index of firms' inflation expectations from earnings call transcripts. This shows departures from a rational framework in firms' inflation expectations and that firms' attention to the central enhances monetary policy effectiveness.
16h00 - Short break
Session 2 – Using NLP to explore different concepts
16h10- Assessing Economic Risks Around Macroeconomic Forecasts: A mixture of fined-tuned BERT and economic experts (Betin, Chalaux, Dex and Turner, OECD, forthcoming)
16h30 - New dimensions of regulatory complexity and their economic cost. An analysis using text mining (De Lucio and Mora-Sanguinetti, Banco de Espana, 2021)
16h50 - Using Computational Linguistics to Identify Competitors and Competitive Interactions (Phillips, Darmouth)
17h10 - Using NLP to detect data/AI hiring intensive jobs and firms (Schmidt, Pilgrim and Mourougane, forthcoming)
17h30 - Closing remarks
Program
