Ei saatavilla suomeksi
Baptiste Meunier
- 13 December 2024
- WORKING PAPER SERIES - No. 3004Details
- Abstract
- We provide a versatile nowcasting toolbox that supports three model classes (dynamic factor models, large Bayesian VAR, bridge equations) and offers methods to manage data selection and adjust for Covid-19 observations. The toolbox aims at simplifying two key tasks: creating new nowcasting models and improving the policy analysis. For model creation, the toolbox automatizes testing input variables, assessing model accuracy, and checking robustness to the Covid period. The toolbox is organized along a structured three-step approach: variable pre-selection, model selection, and Covid robustness. Non-specialists can easily follow these steps to develop high-performing models, while experts can leverage the automated tests and analyses. For regular policy use, the toolbox generates a large range of outputs to aid conjunctural analysis like news decomposition, confidence bands, alternative forecasts, and heatmaps. These multiple outputs aim at opening the "black box" often associated with nowcasts and at gauging the reliability of real-time predictions. We showcase the toolbox features to create a nowcasting model for global GDP growth. Overall, the toolbox aims at facilitating creation, evaluation, and deployment of nowcasting models. Code and templates are available on GitHub: https://github.com/baptiste-meunier/Nowcasting_toolbox.
- JEL Code
- C22 : Mathematical and Quantitative Methods→Single Equation Models, Single Variables→Time-Series Models, Dynamic Quantile Regressions, Dynamic Treatment Effect Models &bull Diffusion Processes
C51 : Mathematical and Quantitative Methods→Econometric Modeling→Model Construction and Estimation
C52 : Mathematical and Quantitative Methods→Econometric Modeling→Model Evaluation, Validation, and Selection
C53 : Mathematical and Quantitative Methods→Econometric Modeling→Forecasting and Prediction Methods, Simulation Methods
C55 : Mathematical and Quantitative Methods→Econometric Modeling→Modeling with Large Data Sets?
- 8 February 2024
- ECONOMIC BULLETIN - BOXEconomic Bulletin Issue 1, 2024Details
- Abstract
- The pandemic triggered the deepest global recession (albeit short-lived) since the Second World War amid large-scale policy support, and led to a sweeping fall in world trade. Following the initial COVID-19 shock, trade staged a rapid recovery, but from the second half of 2022 world trade growth started to decelerate markedly and in 2023 it is estimated to have been considerably below its pre-pandemic average. This box reviews the factors behind the buoyant recovery of global trade following the initial COVID-19 shock and the reasons for its lacklustre performance in 2023, finding that the latter mainly reflects the unwinding of some specific post-pandemic factors (e.g. the rotation of demand from trade-intensive goods towards services owing to the full relaxation of pandemic containment measures) and a less trade-friendly composition of global activity.
- JEL Code
- F01 : International Economics→General→Global Outlook
F1 : International Economics→Trade
F4 : International Economics→Macroeconomic Aspects of International Trade and Finance
- 31 January 2024
- WORKING PAPER SERIES - No. 2900Details
- Abstract
- This paper exploits daily infrared images taken from satellites to track economic activity in advanced and emerging countries. We first develop a framework to read, clean, and exploit satellite images. Our algorithm uses the laws of physics (Planck’s law) and machine learning to detect the heat produced by cement plants in activity. This allows us to monitor in real-time whether a cement plant is working. Using this information on around 500 plants, we construct a satellite-based index tracking activity. We show that using this satellite index outperforms benchmark models and alternative indicators for nowcasting the production of the cement industry as well as the activity in the construction sector. Comparing across methods, we find neural networks yields significantly more accurate predictions as they allow to exploit the granularity of our daily and plant-level data. Overall, we show that combining satellite images and machine learning allows to track economic activity accurately.
- JEL Code
- C51 : Mathematical and Quantitative Methods→Econometric Modeling→Model Construction and Estimation
C81 : Mathematical and Quantitative Methods→Data Collection and Data Estimation Methodology, Computer Programs→Methodology for Collecting, Estimating, and Organizing Microeconomic Data, Data Access
E23 : Macroeconomics and Monetary Economics→Consumption, Saving, Production, Investment, Labor Markets, and Informal Economy→Production
E37 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Forecasting and Simulation: Models and Applications
- 3 August 2023
- WORKING PAPER SERIES - No. 2839Details
- Abstract
- As countries and firms increasingly seek ways to strengthen the resilience of their supply chains, this paper studies the global economic costs of a decoupling of global supply chains along geopolitical lines as well as in strategic sectors. We explore not only the long-run effects, but also the short-run costs stemming from rigid wages and low substitutability across factors of production and input goods. We find that, in terms of welfare losses, the costs of decoupling are roughly five times higher in the short-run compared to the long-run, while country losses are heterogeneous. A reshaping of global supply chains increases the level of consumer prices in most countries, as well as producer prices, especially for trade-intensive manufacturing sectors. Global supply chain decoupling entails also a reallocation of labour across skill levels. Finally, global trade would decrease substantially, driven by lower trade in intermediate inputs and a higher reliance of countries on domestic production.
- JEL Code
- F12 : International Economics→Trade→Models of Trade with Imperfect Competition and Scale Economies, Fragmentation
F13 : International Economics→Trade→Trade Policy, International Trade Organizations
F14 : International Economics→Trade→Empirical Studies of Trade
F51 : International Economics→International Relations, National Security, and International Political Economy→International Conflicts, Negotiations, Sanctions
F62 : International Economics→Economic Impacts of Globalization→Macroeconomic Impacts
- 1 August 2023
- WORKING PAPER SERIES - No. 2836Details
- Abstract
- We nowcast world trade using machine learning, distinguishing between tree-based methods (random forest, gradient boosting) and their regression-based counterparts (macroeconomic random forest, linear gradient boosting). While much less used in the literature, the latter are found to outperform not only the tree-based techniques, but also more “traditional” linear and non-linear techniques (OLS, Markov-switching, quantile regression). They do so significantly and consistently across different horizons and real-time datasets. To further improve performances when forecasting with machine learning, we propose a flexible three-step approach composed of (step 1) pre-selection, (step 2) factor extraction and (step 3) machine learning regression. We find that both pre-selection and factor extraction significantly improve the accuracy of machine-learning-based predictions. This three-step approach also outperforms workhorse benchmarks, such as a PCA-OLS model, an elastic net, or a dynamic factor model. Finally, on top of high accuracy, the approach is flexible and can be extended seamlessly beyond world trade.
- JEL Code
- C53 : Mathematical and Quantitative Methods→Econometric Modeling→Forecasting and Prediction Methods, Simulation Methods
C55 : Mathematical and Quantitative Methods→Econometric Modeling→Modeling with Large Data Sets?
E37 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Forecasting and Simulation: Models and Applications
- 29 March 2023
- ECONOMIC BULLETIN - BOXEconomic Bulletin Issue 2, 2023Details
- Abstract
- This box presents a stylised, model-based, general equilibrium assessment of the global economic effects of trade fragmentation. The focus is on a rather extreme scenario in which two hypothetical geopolitical blocs raise barriers to trade in intermediate goods, causing a relocation of supply chains to countries within the same bloc (“friend-shoring”). Using a model developed by Baqaee and Farhi, we find that economic losses (in terms of welfare, trade and prices) can be sizeable, depending on the degree of rigidities embedded in the model. Effects are also heterogeneous across countries, as small, open economies that are reliant on global value chains are more affected. The findings in this box suggest that trade fragmentation would be a lose-lose situation for all parties involved and leave the global economy more vulnerable to shocks.
- JEL Code
- F12 : International Economics→Trade→Models of Trade with Imperfect Competition and Scale Economies, Fragmentation
F13 : International Economics→Trade→Trade Policy, International Trade Organizations
O33 : Economic Development, Technological Change, and Growth→Technological Change, Research and Development, Intellectual Property Rights→Technological Change: Choices and Consequences, Diffusion Processes
- 7 February 2023
- WORKING PAPER SERIES - No. 2775Details
- Abstract
- We study the effects of negative interest rate policies (NIRP) on the transmission of monetary policy through cross-border lending. Using bank-level data from international financial centres – the United Kingdom, Hong Kong and Ireland – we examine how NIRP in the economies where banks have their headquarters influences cross-border lending from financial-centre affiliates. We find that NIRP impairs the bank-lending channel for cross-border lending to non-bank sectors, especially for those banks that have only a weak deposit base in IFCs – and are thus relatively more exposed to NIRP in their headquarters. Using euro-area data, including bank-level data from France, we find that NIRP does not influence overall cross-border lending from banks’ headquarters’ economies, but NIRP does impair lending to financial sectors based in IFCs. This impairment is stronger for banks with a large deposit base in headquarter economies exposed to NIRP.
- JEL Code
- E52 : Macroeconomics and Monetary Economics→Monetary Policy, Central Banking, and the Supply of Money and Credit→Monetary Policy
F34 : International Economics→International Finance→International Lending and Debt Problems
F36 : International Economics→International Finance→Financial Aspects of Economic Integration
F42 : International Economics→Macroeconomic Aspects of International Trade and Finance→International Policy Coordination and Transmission
G21 : Financial Economics→Financial Institutions and Services→Banks, Depository Institutions, Micro Finance Institutions, Mortgages
- 28 April 2022
- ECONOMIC BULLETIN - BOXEconomic Bulletin Issue 3, 2022Details
- Abstract
- This box reviews the large errors made throughout 2021 and the first quarter of 2022 in Eurosystem and ECB staff inflation projections. Errors in conditioning assumptions, notably due to unexpected energy price increases, are estimated to explain around three-quarters of these errors. Such errors are inherent to the nature of Eurosystem and ECB staff projections, which are conditioned on a set of assumptions, mainly stemming from market-based information including on energy prices. Supply bottlenecks being more persistent than expected, the recovery in economic activity being swifter than predicted, and the transmission of the energy price shock possibly being stronger than usual also played a role, and these factors likely explain a large portion of the errors in projecting HICP inflation excluding energy and food. A comparison with peer institutions shows that large inflation errors were widespread, not only across forecasters but also across economies. This emphasises the predominant role of global factors in a context of steep commodity price increases, especially for energy. While Eurosystem and ECB staff take all available information into account and continuously refine the models used in their projections, inflation developments are likely to remain challenging to forecast in the near term due to the volatile price movements in energy commodities, the uncertainty caused by the war in Ukraine and reopening effects following the removal of pandemic-related restrictions. In this context, complementing the Eurosystem and ECB staff baseline projections with scenario and sensitivity analyses help provide a richer representation of the inflation outlook.
- JEL Code
- C53 : Mathematical and Quantitative Methods→Econometric Modeling→Forecasting and Prediction Methods, Simulation Methods
E37 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Forecasting and Simulation: Models and Applications
E58 : Macroeconomics and Monetary Economics→Monetary Policy, Central Banking, and the Supply of Money and Credit→Central Banks and Their Policies