Pub Date : 2026-01-01Epub Date: 2025-10-09DOI: 10.1016/j.ijforecast.2025.05.003
Laura Coroneo
This paper discusses three key themes in forecasting for monetary policy highlighted in the Bernanke (2024) review: the challenges in economic forecasting, the conditional nature of central bank forecasts, and the importance of forecast evaluation. In addition, a formal evaluation of the Bank of England’s inflation forecasts indicates that, despite the large forecast errors in recent years, they were still accurate relative to common benchmarks.
本文讨论了伯南克（2024）评论中强调的货币政策预测的三个关键主题：经济预测中的挑战、央行预测的条件性质以及预测评估的重要性。此外，对英国央行（Bank of England）通胀预测的正式评估表明，尽管近年来预测误差很大，但相对于普通基准而言，它们仍然是准确的。
{"title":"Forecasting for monetary policy","authors":"Laura Coroneo","doi":"10.1016/j.ijforecast.2025.05.003","DOIUrl":"10.1016/j.ijforecast.2025.05.003","url":null,"abstract":"<div><div>This paper discusses three key themes in forecasting for monetary policy highlighted in the Bernanke (2024) review: the challenges in economic forecasting, the conditional nature of central bank forecasts, and the importance of forecast evaluation. In addition, a formal evaluation of the Bank of England’s inflation forecasts indicates that, despite the large forecast errors in recent years, they were still accurate relative to common benchmarks.</div></div>","PeriodicalId":14061,"journal":{"name":"International Journal of Forecasting","volume":"42 1","pages":"Pages 22-33"},"PeriodicalIF":7.1,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145610005","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01Epub Date: 2025-04-24DOI: 10.1016/j.ijforecast.2025.02.008
Robert W. Rich , Joseph Tracy
Are all forecasters the same? Expectations models incorporating information rigidities typically imply that forecasters are interchangeable, which predicts an absence of systematic patterns in individual forecast behavior. Motivated by this prediction, we examine the European Central Bank’s Survey of Professional Forecasters and find, in contrast, that participants display systematic patterns in predictive performance both within and across target variables. Moreover, we document a new result from professional forecast surveys, which is that inter- and intra-forecaster relative predictive performance are strongly linked to the degree of difficulty in the forecasting environment. This insight can inform the ongoing development of expectations models.
{"title":"All forecasters are not the same: Systematic patterns in predictive performance","authors":"Robert W. Rich , Joseph Tracy","doi":"10.1016/j.ijforecast.2025.02.008","DOIUrl":"10.1016/j.ijforecast.2025.02.008","url":null,"abstract":"<div><div>Are all forecasters the same? Expectations models incorporating information rigidities typically imply that forecasters are interchangeable, which predicts an absence of systematic patterns in individual forecast behavior. Motivated by this prediction, we examine the European Central Bank’s Survey of Professional Forecasters and find, in contrast, that participants display systematic patterns in predictive performance both within and across target variables. Moreover, we document a new result from professional forecast surveys, which is that inter- and intra-forecaster relative predictive performance are strongly linked to the degree of difficulty in the forecasting environment. This insight can inform the ongoing development of expectations models.</div></div>","PeriodicalId":14061,"journal":{"name":"International Journal of Forecasting","volume":"42 1","pages":"Pages 235-258"},"PeriodicalIF":7.1,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145610317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01Epub Date: 2025-11-19DOI: 10.1016/j.ijforecast.2025.11.001
Nikolaos Kourentzes , Robert Fildes
We complement the previous discussions of Bernanke’s review of the Bank of England’s forecasting activities and highlight directions for future research that are relevant to central banks and the wider forecasting community. Decisions in central banks, such as monetary policy ones, are hardly algorithmic and are often influenced by policy and current soft contextual information, introducing challenges into evaluating and specifying forecasts. The use of alternatives to standard econometric models is highlighted in the Bernanke report and other commentaries in this series. These methodological alternatives require both more research, to be validly applied and evaluated, and a cultural shift for those with forecasting responsibilities in central banks. Critically, uncertainty estimates in central bank forecasts are hardly purely model-based. How this is done and how to best communicate it to stakeholders and counterparties are fertile areas for research with potentially important implications for market participants. Finally, while academic research often focuses on large, well-funded central banks, there is a significant opportunity to help smaller, less-resourced institutions.
{"title":"Beyond the numbers: The role of people and processes in central bank forecasting","authors":"Nikolaos Kourentzes , Robert Fildes","doi":"10.1016/j.ijforecast.2025.11.001","DOIUrl":"10.1016/j.ijforecast.2025.11.001","url":null,"abstract":"<div><div>We complement the previous discussions of Bernanke’s review of the Bank of England’s forecasting activities and highlight directions for future research that are relevant to central banks and the wider forecasting community. Decisions in central banks, such as monetary policy ones, are hardly algorithmic and are often influenced by policy and current soft contextual information, introducing challenges into evaluating and specifying forecasts. The use of alternatives to standard econometric models is highlighted in the Bernanke report and other commentaries in this series. These methodological alternatives require both more research, to be validly applied and evaluated, and a cultural shift for those with forecasting responsibilities in central banks. Critically, uncertainty estimates in central bank forecasts are hardly purely model-based. How this is done and how to best communicate it to stakeholders and counterparties are fertile areas for research with potentially important implications for market participants. Finally, while academic research often focuses on large, well-funded central banks, there is a significant opportunity to help smaller, less-resourced institutions.</div></div>","PeriodicalId":14061,"journal":{"name":"International Journal of Forecasting","volume":"42 1","pages":"Pages 40-43"},"PeriodicalIF":7.1,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145610398","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01Epub Date: 2025-08-09DOI: 10.1016/j.ijforecast.2025.07.003
David Ardia , Keven Bluteau
We propose an approach to construct text-based time-series indices in an optimal way—typically, indices that maximize the contemporaneous relation or the predictive performance with respect to a target variable, such as inflation. Our methodology relies on binary selection matrices that, applied to the vocabulary of tokens, select the relevant texts in the corpus. Various widely known text-based indices, such as the Economic Policy Uncertainty (EPU) index, can be formulated in terms of selection matrices. We design a genetic algorithm with domain-specific knowledge featuring tailor-made crossover and mutation operations to perform the complex optimization. We illustrate our methodology with a corpus of news articles from the Wall Street Journal by optimizing text-based indices that forecast inflation at various horizons.
我们提出了一种以最优方式构建基于文本的时间序列指数的方法-通常，指数最大化同期关系或相对于目标变量（如通货膨胀）的预测性能。我们的方法依赖于二进制选择矩阵，应用于标记的词汇表，选择语料库中的相关文本。各种众所周知的基于文本的指数，如经济政策不确定性（EPU）指数，都可以根据选择矩阵来制定。我们设计了一种具有特定领域知识的遗传算法，该算法具有定制的交叉和突变操作来执行复杂的优化。我们以《华尔街日报》（Wall Street Journal）的大量新闻文章为例，通过优化基于文本的指数来说明我们的方法，这些指数可以预测不同时期的通货膨胀。
{"title":"Optimal text-based time-series indices","authors":"David Ardia , Keven Bluteau","doi":"10.1016/j.ijforecast.2025.07.003","DOIUrl":"10.1016/j.ijforecast.2025.07.003","url":null,"abstract":"<div><div>We propose an approach to construct text-based time-series indices in an optimal way—typically, indices that maximize the contemporaneous relation or the predictive performance with respect to a target variable, such as inflation. Our methodology relies on binary selection matrices that, applied to the vocabulary of tokens, select the relevant texts in the corpus. Various widely known text-based indices, such as the Economic Policy Uncertainty (EPU) index, can be formulated in terms of selection matrices. We design a genetic algorithm with domain-specific knowledge featuring tailor-made crossover and mutation operations to perform the complex optimization. We illustrate our methodology with a corpus of news articles from the <em>Wall Street Journal</em> by optimizing text-based indices that forecast inflation at various horizons.</div></div>","PeriodicalId":14061,"journal":{"name":"International Journal of Forecasting","volume":"42 1","pages":"Pages 44-60"},"PeriodicalIF":7.1,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145610399","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01Epub Date: 2025-05-19DOI: 10.1016/j.ijforecast.2025.03.006
David Aikman , Richard Barwell
We summarize reactions to the Bernanke Review from the Bank of England watchers community – a diverse group of academics, market economists, and business analysts who closely monitor and analyze the actions of the Bank of England. Key themes include the Review’s recommendations to retire the “fan chart”, increase the use of scenario analysis, and de-emphasize the central forecast conditioned on the market yield curve, as well as its critique of the Bank’s forecasting infrastructure. There is also extensive discussion of areas left unaddressed by the Review, including whether the Monetary Policy Committee should publish its preferred policy rate path, adopt a Fed-style dot plot, or give Bank staff ownership of the forecast.
{"title":"Reactions to the Bernanke Review from Bank of England watchers","authors":"David Aikman , Richard Barwell","doi":"10.1016/j.ijforecast.2025.03.006","DOIUrl":"10.1016/j.ijforecast.2025.03.006","url":null,"abstract":"<div><div>We summarize reactions to the Bernanke Review from the Bank of England watchers community – a diverse group of academics, market economists, and business analysts who closely monitor and analyze the actions of the Bank of England. Key themes include the Review’s recommendations to retire the “fan chart”, increase the use of scenario analysis, and de-emphasize the central forecast conditioned on the market yield curve, as well as its critique of the Bank’s forecasting infrastructure. There is also extensive discussion of areas left unaddressed by the Review, including whether the Monetary Policy Committee should publish its preferred policy rate path, adopt a Fed-style dot plot, or give Bank staff ownership of the forecast.</div></div>","PeriodicalId":14061,"journal":{"name":"International Journal of Forecasting","volume":"42 1","pages":"Pages 3-12"},"PeriodicalIF":7.1,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145610395","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01Epub Date: 2025-10-08DOI: 10.1016/j.ijforecast.2025.07.001
Jennifer L. Castle , Jurgen A. Doornik , David F. Hendry
The Bank of England badly mis-forecast UK annual consumer price inflation as it rose rapidly from 2021, prompting a review by Ben Bernanke. This raised many important issues, but other crucial problems were not addressed, as we discuss. Unpredictable shocks explain some of the bank’s forecast failures, but tardy reactions also mattered. We show that successive large and increasing same-sign one-step-ahead forecast errors contain the information to estimate broken trends, applied to forecasting the UK’s inflation over 2021–24. Compared with Bank of England projections, substantial gains in forecast performance can be made by rapidly detecting trend breaks and updating forecasting models when they occur.
英国央行（Bank of England）严重错误地预测了英国消费者价格指数（cpi）从2021年开始迅速上升，促使本•伯南克（Ben Bernanke）重新审视。这提出了许多重要的问题，但其他关键问题没有得到解决，正如我们讨论的那样。不可预测的冲击解释了该行预测失败的部分原因，但反应迟缓也很重要。我们表明，连续的大且不断增加的同号一步预测误差包含了估计断裂趋势的信息，应用于预测英国2021-24年的通货膨胀。与英国央行（Bank of England）的预测相比，通过快速发现趋势突变并在突变发生时更新预测模型，预测业绩可以大幅提升。
{"title":"Could the Bank of England have avoided mis-forecasting UK inflation during 2021–24?","authors":"Jennifer L. Castle , Jurgen A. Doornik , David F. Hendry","doi":"10.1016/j.ijforecast.2025.07.001","DOIUrl":"10.1016/j.ijforecast.2025.07.001","url":null,"abstract":"<div><div>The Bank of England badly mis-forecast UK annual consumer price inflation as it rose rapidly from 2021, prompting a review by Ben Bernanke. This raised many important issues, but other crucial problems were not addressed, as we discuss. Unpredictable shocks explain some of the bank’s forecast failures, but tardy reactions also mattered. We show that successive large and increasing same-sign one-step-ahead forecast errors contain the information to estimate broken trends, applied to forecasting the UK’s inflation over 2021–24. Compared with Bank of England projections, substantial gains in forecast performance can be made by rapidly detecting trend breaks and updating forecasting models when they occur.</div></div>","PeriodicalId":14061,"journal":{"name":"International Journal of Forecasting","volume":"42 1","pages":"Pages 13-21"},"PeriodicalIF":7.1,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145610396","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01Epub Date: 2025-06-30DOI: 10.1016/j.ijforecast.2025.05.001
Xiuqin Xu, Hanqiu Peng, Ying Chen
Modern time series data often display complex nonlinear dependencies along with irregular regime-switching behaviors. These features present technical challenges in modeling, inference, and providing insightful understanding of the underlying stochastic phenomena. To tackle these challenges, we introduce the novel Deep Switching State Space Model (DSM). In DSM, the architecture employs discrete latent variables to represent regimes and continuous latent variables to account for random driving factors. By melding a Recurrent Neural Network (RNN) with a nonlinear Switching State Space Model (SSSM), we manage to capture the nonlinear dependencies and irregular regime-switching behaviors, governed by a Markov chain and parameterized using multilayer perceptrons. We validate the DSM through short- and long-term forecasting on a wide array of simulated and real-world datasets, spanning sectors such as healthcare, economics, traffic, meteorology, and energy. Our results reveal that DSM outperforms several state-of-the-art models in terms of forecasting accuracy, while providing meaningful regime identifications.
{"title":"Deep switching state space model for nonlinear time series forecasting with regime switching","authors":"Xiuqin Xu, Hanqiu Peng, Ying Chen","doi":"10.1016/j.ijforecast.2025.05.001","DOIUrl":"10.1016/j.ijforecast.2025.05.001","url":null,"abstract":"<div><div>Modern time series data often display complex nonlinear dependencies along with irregular regime-switching behaviors. These features present technical challenges in modeling, inference, and providing insightful understanding of the underlying stochastic phenomena. To tackle these challenges, we introduce the novel Deep Switching State Space Model (DS<span><math><msup><mrow></mrow><mrow><mn>3</mn></mrow></msup></math></span>M). In DS<span><math><msup><mrow></mrow><mrow><mn>3</mn></mrow></msup></math></span><span>M, the architecture employs discrete latent variables to represent regimes and continuous latent variables to account for random driving factors. By melding a Recurrent Neural Network<span> (RNN) with a nonlinear Switching State Space Model (SSSM), we manage to capture the nonlinear dependencies and irregular regime-switching behaviors, governed by a Markov chain<span> and parameterized using multilayer perceptrons. We validate the DS</span></span></span><span><math><msup><mrow></mrow><mrow><mn>3</mn></mrow></msup></math></span>M through short- and long-term forecasting on a wide array of simulated and real-world datasets, spanning sectors such as healthcare, economics, traffic, meteorology, and energy. Our results reveal that DS<span><math><msup><mrow></mrow><mrow><mn>3</mn></mrow></msup></math></span>M outperforms several state-of-the-art models in terms of forecasting accuracy, while providing meaningful regime identifications.</div></div>","PeriodicalId":14061,"journal":{"name":"International Journal of Forecasting","volume":"42 1","pages":"Pages 85-98"},"PeriodicalIF":7.1,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145610401","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01Epub Date: 2025-04-30DOI: 10.1016/j.ijforecast.2025.03.007
Arndt Leininger , Andreas E. Murr , Lukas Stötzer , Mark A. Kayser
Existing studies show that aggregating citizens’ expectations about who will win can predict election outcomes in a majoritarian system. But can so-called citizen forecasting also successfully predict outcomes in mixed-member systems, where constituency results are less important? The existing evidence is mixed and limited in scope. We conducted, therefore, a citizen forecast of the 2021 German federal election by administering an original survey asking citizens who they thought would win in their constituency, what share of the vote each candidate would win in their constituency, and what share of the vote each party would win nationally. Citizens predicted constituency winners and vote shares more accurately than several benchmarks. However, our citizen forecast was based on a non-representative sample from an online-access panel. We conclude that citizen forecasting provides a simple and inexpensive way to predict the various relevant outcomes in mixed-member elections.
{"title":"Citizen forecasting in a mixed electoral system","authors":"Arndt Leininger , Andreas E. Murr , Lukas Stötzer , Mark A. Kayser","doi":"10.1016/j.ijforecast.2025.03.007","DOIUrl":"10.1016/j.ijforecast.2025.03.007","url":null,"abstract":"<div><div>Existing studies show that aggregating citizens’ expectations about who will win can predict election outcomes in a majoritarian system. But can so-called citizen forecasting also successfully predict outcomes in mixed-member systems, where constituency results are less important? The existing evidence is mixed and limited in scope. We conducted, therefore, a citizen forecast of the 2021 German federal election by administering an original survey asking citizens who they thought would win in their constituency, what share of the vote each candidate would win in their constituency, and what share of the vote each party would win nationally. Citizens predicted constituency winners and vote shares more accurately than several benchmarks. However, our citizen forecast was based on a non-representative sample from an online-access panel. We conclude that citizen forecasting provides a simple and inexpensive way to predict the various relevant outcomes in mixed-member elections.</div></div>","PeriodicalId":14061,"journal":{"name":"International Journal of Forecasting","volume":"42 1","pages":"Pages 203-215"},"PeriodicalIF":7.1,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145610316","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01Epub Date: 2025-04-16DOI: 10.1016/j.ijforecast.2025.03.004
Tim J. Boonen, Yuhuai Chen
We introduce a spatial–temporally weighted vector autoregressive (SWVAR) model for modeling and forecasting mortality rates across multiple populations. First, we stack the mortality rates of the populations and build a vector autoregressive (VAR) model. Next, we apply the sparse group least absolute shrinkage and selection operator (sparse group LASSO) for fitting to avoid overparameterization. Furthermore, we integrate spatial–temporal weights, derived from age differences and geographic centroid distances, into the grouped penalty term. These approaches allow the resulting model to effectively combine information from multiple populations and reduce confounding factors associated with combined modeling. We demonstrate through a series of empirical experiments that the spatial–temporally weighted VAR model enhances estimation accuracy and exhibits superior in-sample fitting and out-of-sample forecasting performance.
{"title":"VAR Model with Sparse Group LASSO for Multi-population Mortality Forecasting","authors":"Tim J. Boonen, Yuhuai Chen","doi":"10.1016/j.ijforecast.2025.03.004","DOIUrl":"10.1016/j.ijforecast.2025.03.004","url":null,"abstract":"<div><div>We introduce a spatial–temporally weighted vector autoregressive (SWVAR) model for modeling and forecasting mortality rates across multiple populations. First, we stack the mortality rates of the populations and build a vector autoregressive (VAR) model. Next, we apply the sparse group least absolute shrinkage and selection operator (sparse group LASSO) for fitting to avoid overparameterization. Furthermore, we integrate spatial–temporal weights, derived from age differences and geographic centroid distances, into the grouped penalty term. These approaches allow the resulting model to effectively combine information from multiple populations and reduce confounding factors associated with combined modeling. We demonstrate through a series of empirical experiments that the spatial–temporally weighted VAR model enhances estimation accuracy and exhibits superior in-sample fitting and out-of-sample forecasting performance.</div></div>","PeriodicalId":14061,"journal":{"name":"International Journal of Forecasting","volume":"42 1","pages":"Pages 259-280"},"PeriodicalIF":7.1,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145610318","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01Epub Date: 2025-05-22DOI: 10.1016/j.ijforecast.2025.04.006
Nicolas Rost , Michele Ronco
This exploratory study assesses the risk of future onset of large-scale, conflict-related internal displacement in countries facing humanitarian emergencies. We train a variety of machine learning models on near-real-time data, which we compare against a simple baseline model, to assess the risk, one and three months into the future, of whether at least 1,000 people per month will flee their homes due to conflict. Measures of past displacement, conflict, risk of humanitarian crises, humanitarian access, the severity of humanitarian crises, and free elections improve forecasting performance. Limitations include the fact that displacement onsets are rare and hard to predict, and limited data availability and quality. Still, the best random forest model flagged 24 of 26 cases of displacement onset three months into the future and identified a high-risk group of country-months with a 33 times higher probability of displacement onset than a low-risk group. Providing such monthly forecasts to humanitarian practitioners could help them prepare better for new displacement or even mitigate the human suffering caused by conflict.
{"title":"Anticipating humanitarian emergencies with a high risk of conflict-induced displacement","authors":"Nicolas Rost , Michele Ronco","doi":"10.1016/j.ijforecast.2025.04.006","DOIUrl":"10.1016/j.ijforecast.2025.04.006","url":null,"abstract":"<div><div>This exploratory study assesses the risk of future onset of large-scale, conflict-related internal displacement in countries facing humanitarian emergencies. We train a variety of machine learning models on near-real-time data, which we compare against a simple baseline model, to assess the risk, one and three months into the future, of whether at least 1,000 people per month will flee their homes due to conflict. Measures of past displacement, conflict, risk of humanitarian crises, humanitarian access, the severity of humanitarian crises, and free elections improve forecasting performance. Limitations include the fact that displacement onsets are rare and hard to predict, and limited data availability and quality. Still, the best random forest model flagged 24 of 26 cases of displacement onset three months into the future and identified a high-risk group of country-months with a 33 times higher probability of displacement onset than a low-risk group. Providing such monthly forecasts to humanitarian practitioners could help them prepare better for new displacement or even mitigate the human suffering caused by conflict.</div></div>","PeriodicalId":14061,"journal":{"name":"International Journal of Forecasting","volume":"42 1","pages":"Pages 138-157"},"PeriodicalIF":7.1,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145610405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号