Publication: Comparative Analysis of AI-Predicted and Crowdsourced Food Prices in an Economically Volatile Region
Loading...
Date
2024-04-23
ISSN
Published
2024-04-23
Editor(s)
Abstract
High-frequency monitoring of food commodity prices is important for assessing and responding to shocks, especially in fragile contexts where timely and targeted interventions for food security are critical. However, national price surveys are typically limited in temporal and spatial granularity. It is cost prohibitive to implement traditional data collection at frequent timescales to unravel spatiotemporal price evolution across market segments and at subnational geographic levels. Recent advancements in data innovation offer promising solutions to address the paucity of commodity price data and guide market intelligence for diverse development stakeholders. The use of artificial intelligence to estimate missing price data and a parallel effort to crowdsource commodity price data are both unlocking cost-effective opportunities to generate actionable price data. Yet, little is known about how the data from these alternative methods relate to independent ground truth data. To evaluate if these data strategies can meet the long-standing demand for real-time intelligence on food affordability, this paper analyzes open-source daily crowdsourced data (104,931 datapoints) from a recently published data set in Nature Journal, relative to complementary ground truth sample. The paper subsequently compares these data to open-source monthly artificial intelligence–generated price data for identical commodities over a 36-month period in northern Nigeria, from 2019 to 2022. The results show that all the data sources share a high degree of comparability, with variation across commodity and market segments. Overall, the findings provide important support for leveraging these new and innovative data approaches to enable data-driven decision-making in near real time.
Link to Data Set
Citation
“Adewopo, Julius; Andrée, Bo Pieter Johannes; Peter, Helen; Solano-Hermosilla, Gloria; Micale, Fabio. 2024. Comparative Analysis of AI-Predicted and Crowdsourced Food Prices in an Economically Volatile Region. Policy Research Working Paper; 10758. © World Bank. http://hdl.handle.net/10986/41467 License: CC BY 3.0 IGO.”
Associated URLs
Associated content
Other publications in this report series
Publication Dynamic, High-Resolution Wealth Measurement in Data-Scarce Environments(Washington, DC: World Bank, 2025-02-06)Accurate and comprehensive measurement of household livelihoods is critical for monitoring progress toward poverty alleviation and targeting social assistance programs for those who most need it. However, the high cost of traditional data collection has historically made comprehensive measurement a difficult task. This paper evaluates alternative satellite-based deep learning approaches using detailed household census extracts from four African countries to accelerate progress toward comprehensive, fine-scale, and dynamic measurement of asset wealth at scale. The results indicate that transformer architectures solve multiple open measurement problems, by providing the most accurate measurement of local-level variation in household asset wealth across countries and cities, as well as changes in household asset wealth over time. Experiments that artificially restrict data availability show the model’s ability to achieve high performance with limited data. The proposed approach demonstrates the promise of combining satellite imagery, publicly available geo-features, and new deep learning architectures for hyperlocal and dynamic measurement of wealth in data-scarce environments.Publication Firm-Level Climate Change Adaptation(Washington, DC: World Bank, 2025-03-10)Are firms adapting to climate change? This paper studies this question by combining geocoded World Bank Enterprise Survey data with spatially granular weather data to estimate temperature response functions for nearly 160,000 firms in 134 countries over a 15-year period. Our results show that market imperfections in low- and middle-income countries constrain firms’ ability to adapt. Small and medium-size firms in low- and low-middle income countries are most vulnerable, with revenues declining by 12 percent in years with temperatures 0.5◦C above historical averages. The impact is equally strong for manufacturing and services firms and result from declines in labor productivity and wages. Heat-sensitive sectors and less resilient firms are more severely affected, reinforcing the causal interpretation. Unique firm-level information on policy constraints including limited financing, burdensome regulations, and unsafe conditions suggest that such factors raise adaptation costs, undermining economic resilience to climate change.Publication Beyond the AI Divide(Washington, DC: World Bank, 2025-02-24)This paper examines global disparities in artificial intelligence preparedness, using the 2023 Artificial Intelligence Preparedness Index developed by the International Monetary Fund alongside the multidimensional Economic Complexity Index. The proposed methodology identifies both global and local overperformers by comparing actual artificial intelligence readiness scores to predictions based on economic complexity, offering a comprehensive assessment of national artificial intelligence capabilities. The findings highlight the varying significance of regulation and ethics frameworks, digital infrastructure, as well as human capital and labor market development in driving artificial intelligence overperformance across different income levels. Through case studies, including Singapore, Northern Europe, Malaysia, Kazakhstan, Ghana, Rwanda, and emerging demographic giants like China and India, the analysis illustrates how even resource-constrained nations can achieve substantial artificial intelligence advancements through strategic investments and coherent policies. The study underscores the need for offering actionable insights to foster peer learning and knowledge-sharing among countries. It concludes with recommendations for improving artificial intelligence preparedness metrics and calls for future research to incorporate cognitive and cultural dimensions into readiness frameworks.Publication Indigenous peoples, land and conflict in Mindanao, Philippines(Washington, DC: World Bank, 2024-02-12)This article explores the links between conflict, land and indigenous peoples in several regions of Mindano, the Philippines, notorious for their levels of poverty and conflict. The analysis takes advantage of the unprecedented concurrence of data from the most recent, 2020, census; an independent conflict data monitor for Mindanao; and administrative sources on ancestral land titling for indigenous peoples in the Philippines. While evidence elsewhere compellingly links land titling with conflict reduction, a more nuanced story emerges in the Philippines. Conflicts, including land- and resource-related conflicts, are generally less likely in districts (barangays) with higher shares of indigenous peoples. Ancestral domain areas also have a lower likelihood for general conflict but a higher likelihood for land-related conflict. Ancestral domains titling does not automatically solve land-related conflicts. When administrative delays take place (from cumbersome bureaucratic processes, insufficient resources and weak institutional capacity), titling processes may lead to sustained, rather than decreased, conflict.Publication Who on Earth Is Using Generative AI ?(Washington, DC: World Bank, 2024-08-22)Leveraging unconventional data, including website traffic data and Google Trends, this paper unveils the real-time usage patterns of generative artificial intelligence tools by individuals across countries. The paper also examines country-level factors driving the uptake and early impacts of generative artificial intelligence on online activities. As of March 2024, the top 40 generative artificial intelligence tools attract nearly 3 billion visits per month from hundreds of millions of users. ChatGPT alone commanded 82.5 percent of the traffic, yet reaching only one-eightieth of Google’s monthly visits. Generative artificial intelligence users skew young, highly educated, and male, particularly for video generation tools, with usage patterns strongly indicating productivity-related activities. Generative artificial intelligence has achieved unprecedentedly rapid global diffusion, reaching almost all economies worldwide within 16 months of ChatGPT’s release. Middle-income economies have disproportionately high adoption of generative artificial intelligence relative to their economic scale, now contribute more than 50 percent of global traffic, while low-income economies contribute less than 1 percent. Regression analysis reveals that income level, share of youth population, digital infrastructure, specialization in high-skill tradable services, English proficiency, and human capital are strongly correlated with higher uptake of generative artificial intelligence. The paper also documents disruptions in online traffic patterns and emphasizes the need for targeted investments in digital infrastructure and skills development to harness the full potential of artificial intelligence.
Journal
Journal Volume
Journal Issue
Citations
Collections
Related items
Showing items related by metadata.
Publication A Data-Driven Approach for Early Detection of Food Insecurity in Yemen's Humanitarian Crisis(Washington, DC: World Bank, 2024-05-09)The Republic of Yemen is enduring the world's most severe protracted humanitarian crisis, compounded by conflict, economic collapse, and natural disasters. Current food insecurity assessments rely on expert evaluation of evidence with limited temporal frequency and foresight. This paper introduces a data-driven methodology for the early detection and diagnosis of food security emergencies. The approach optimizes for simplicity and transparency, and pairs quantitative indicators with data-driven optimal thresholds to generate early warnings of impending food security emergencies. Historical validation demonstrates that warnings can be reliably issued before sharp deterioration in food security occurs, using only a few critical indicators that capture inflation, conflict, and agricultural productivity shocks. These indicators signal deterioration most accurately at five months of lead time. The paper concludes that simple data-driven approaches show a strong capability to generate reliable food security warnings in Yemen, highlighting their potential to complement existing assessments and enhance lead time for effective intervention.Publication Altered Destinies: The Long-Term Effects of Rising Prices and Food Insecurity in the Middle East and North Africa(Washington, DC : World Bank, 2023-04-06)Growth is forecasted to slow down for the Middle East and North Africa region. The war in Ukraine in 2022 exacerbated inflationary pressures as the world recovered from the COVID 19 pandemic induced recession. The response by central banks to raise rates to curb inflation is slowing economic activity, while rising food prices are making it difficult for families to put meals on the table. Inflation, when it stems from food prices, hits the poor harder than the rich, thus compounding food insecurity in MENA that had been rising over decades. The immediate effects of food insecurity can be a devastating loss of life, but even temporary increases in food prices can cause long-term irreversible damages, especially to children. The rise in food prices due to the war in Ukraine may have altered the destinies of hundreds of thousands of children in the region, setting them on paths to limited prosperity. Food insecurity imposes challenges to a region where the state of child nutrition and health were inadequate before the shocks from the COVID-19 pandemic. The report discusses policy options and highlights the need for data to guide effective decision making.Publication Estimating Food Price Inflation from Partial Surveys(World Bank, Washington, DC, 2021-12)The traditional consumer price index is often produced at an aggregate level, using data from few, highly urbanized, areas. As such, it poorly describes price trends in rural or poverty-stricken areas, where large populations may reside in fragile situations. Traditional price data collection also follows a deliberate sampling and measurement process that is not well suited for monitoring during crisis situations, when price stability may deteriorate rapidly. To gain real-time insights beyond what can be formally measured by traditional methods, this paper develops a machine-learning approach for imputation of ongoing subnational price surveys. The aim is to monitor inflation at the market level, relying only on incomplete and intermittent survey data. The capabilities are highlighted using World Food Programme surveys in 25 fragile and conflict-affected countries where real-time monthly food price data are not publicly available from official sources. The results are made available as a data set that covers more than 1200 markets and 43 food types. The local statistics provide a new granular view on important inflation events, including the World Food Price Crisis of 2007–08 and the surge in global inflation following the 2020 pandemic. The paper finds that imputations often achieve accuracy similar to direct measurement of prices. The estimates may provide new opportunities to investigate local price dynamics in markets where prices are sensitive to localized shocks and traditional data are not available.Publication Predicting Food Crises(World Bank, Washington, DC, 2020-09)Globally, more than 130 million people are estimated to be in food crisis. These humanitarian disasters are associated with severe impacts on livelihoods that can reverse years of development gains. The existing outlooks of crisis-affected populations rely on expert assessment of evidence and are limited in their temporal frequency and ability to look beyond several months. This paper presents a statistical forecasting approach to predict the outbreak of food crises with sufficient lead time for preventive action. Different use cases are explored related to possible alternative targeting policies and the levels at which finance is typically unlocked. The results indicate that, particularly at longer forecasting horizons, the statistical predictions compare favorably to expert-based outlooks. The paper concludes that statistical models demonstrate good ability to detect future outbreaks of food crises and that using statistical forecasting approaches may help increase lead time for action.Publication Climate Shocks and Their Effects on Food Security, Prices, and Agricultural Wages in Afghanistan(Washington, DC: World Bank, 2024-12-17)This study examines the effects of climate and weather shocks on Afghanistan's agricultural economy, with an emphasis on food security, prices, and wages. By utilizing a dynamical model and a unique data set that includes monthly global and local food prices, agricultural wages, unofficial exchange rates, and local climate data, the research provides econometric estimates of the impacts of droughts and floods. The findings reveal that both flooding and drought significantly increase food insecurity, directly and indirectly. Directly, these climatic shocks are linked to heightened risks of food insecurity in the following months, even when controlling for price and wage fluctuations. Indirectly, droughts and floods drive up food prices and depress agricultural wages, further exacerbating food insecurity. The study suggests that enhancing climate resilience in the agriculture sector could mitigate these risks, stabilize local food prices and wages, and strengthen food security and the broader agricultural economy. The results also show that price data effectively capture food security shocks from various non-economic sources, and can serve as a versatile monitoring tool in situations where detailed data on food security are unavailable.
Users also downloaded
Showing related downloaded files
Publication Macroeconomic and Fiscal Implications of Population Aging in Bulgaria(World Bank, Washington, DC, 2014-02)Bulgaria is in the midst of a serious demographic transition that will shrink its population at one of the highest rates in the world within the next few decades. This study analyzes the macroeconomic and fiscal implications of this demographic transition by using a long-term model, which integrates the demographic projections with social security, fiscal and real economy dimensions in a consistent manner. The simulations suggest that, even under fairly optimistic assumptions, Bulgaria's demographic transition will exert significant fiscal pressures and depress the economic growth in the medium and long term. However, the results also demonstrate that the Government of Bulgaria can play a significant role in mitigating some of these effects. Policies that induce higher labor force participation, promote productivity and technological improvement, and provide better education outcomes are found to counteract the negative consequences of the demographic shift.Publication Environmental Flows in Water Resources Policies, Plans, and Projects : Findings and Recommendations(World Bank, 2009)The overall goal of the analysis presented in this report is to advance the understanding and integration in operational terms of environmental water allocation into integrated water resources management. The specific objectives of this report are the following: 1) document the changing understanding of environmental flows, by both water resources practitioners and by environmental experts within the Bank and in borrowing countries; 2) draw lessons from experience in implementing environmental flows by the Bank, other international development organizations with experience in this area, and a small number of developed and developing countries; 3) develop an analytical framework to support more effective integration of environmental flow considerations for informing and guiding: (a) the planning, design, and operations decision making of water resources infrastructure projects; (b) the legal, policy, institutional, and capacity development related to environmental flows; and (c) restoration programs; and 4) provide recommendations for improvements in technical guidance to better incorporate environmental flow considerations into the preparation and implementation of lending operations.Publication World Development Indicators 2010(World Bank, 2010-04-01)The 1998 edition of world development indicators initiated a series of annual reports on progress toward the International development goals. In the foreword then, World Bank President James D. Wolfensohn recognized that 'by reporting regularly and systematically on progress toward the targets the international community has set for itself, the author will focus attention on the task ahead and make those responsible for advancing the development agenda accountable for results.' The same vision inspired world leaders to commit themselves to the millennium development goals. On this, the 10th anniversary of the millennium declaration, world development indicators 2010 focuses on progress toward the millennium development goals and the challenges of meeting them.Publication Opportunity Assessment to Strengthen Collective Land Tenure Rights in FCPF Countries(World Bank, Washington, DC, 2021-10)Governments, development institutions, and the private sector are increasingly turning to nature-based solutions to address the world’s climate and biodiversity crisis. Countries, corporations, and investors are increasingly looking to forest- and land-based emission reduction programs (ERPs) to achieve early mitigation gains while they develop longer-term strategies and solutions to cut their greenhouse gas emissions. Central to emerging natural climate solutions are efforts to reduce deforestation and forest degradation while encouraging restoration, conservation, and sustainable use of forests in developing countries. The Forest Carbon Partnership Facility (FCPF), which became operational in June 2008, is a global partnership focused on reducing emissions from deforestation and forest degradation, forest carbon stock conservation, the sustainable management of forests and enhancement of forest carbon stocks (REDD+). Communal land and forest tenure rights for Indigenous Peoples and local communities (IPLCs) is critical for the success of emission reduction program (ERP) implementation. The remainder of this report is structured as follows. Section 2 provides an overview of the analytical and methodological approach of the study. Section 3 discusses core findings about the nature and range of emergent opportunities associated with efforts to advance, strengthen, and leverage rights and presents the main opportunities in six selected countries. Section 4 discusses lessons learned and cross-cutting areas for further development of rights recognition as a global process. Section 5 provides a summary of the country profiles.Publication Stolen Asset Recovery : A Good Practices Guide for Non-conviction Based Asset Forfeiture(World Bank, 2009)The guide is organized into three major parts: Part A first provides an overview of the problem of stolen assets and the problem of recovering the assets once they are transferred abroad. Second, it describes how the international community has taken steps to respond to the problem through United Nations Convention against Corruption (UNCAC) and the Stolen Asset Recovery (StAR) Initiative. UNCAC introduced a new framework to facilitate the tracing, freezing, seizing, forfeiture, and return of assets stolen through corrupt practices and hidden in foreign jurisdictions. The StAR Initiative developed an action plan to support the domestication and implementation of asset recovery provisions under UNCAC, to facilitate countries' efforts to recover stolen assets that have been hidden in foreign jurisdictions, and ultimately, to help deter such flows and eliminate safe havens for hiding corruption proceeds. Third and finally, Part A introduces non-conviction based (NCB) asset forfeiture as one of the critical tools to combat corruption, describing the situations when it is useful, how it differs from criminal forfeiture, its usefulness in civil and common law jurisdictions, and the support it has gained internationally. Part B contains the 36 key concepts. The concepts have been grouped together by topic area, including prime imperatives, definitions of assets and offenses subject to NCB asset forfeiture, measures for investigation and preservation of assets, procedural and evidentiary concepts, determining parties and ensuring proper notice, judgment proceedings, organizational considerations and asset management, and international cooperation and asset recovery. The concepts are illustrated through examples from cases and excerpts from different jurisdictions' NCB asset forfeiture legislation. Part C contains a number of special contributions written by individual practitioners. The contributions focus on the general practice of NCB asset forfeiture and international cooperation in specific jurisdictions, namely Colombia, Guernsey, Ireland, Kuwait, Switzerland, Thailand, and the United Kingdom. In addition, some contributions illustrate a selection of NCB asset forfeiture practices, such as asset management, delegating certain roles to the executive branch, and pursuing forfeiture based on illicit enrichment.