{
  "title": "Nowcasting Economic Growth with Machine Learning and Satellite Data",
  "publication": "IMF Working Papers, January 30, 2026",
  "sourceUrl": "https://www.imf.org/en/publications/wp/issues/2026/01/30/nowcasting-economic-growth-with-machine-learning-and-satellite-data-573623",
  "canonical": "https://www.imf.org/en/publications/wp/issues/2026/01/30/nowcasting-economic-growth-with-machine-learning-and-satellite-data-573623",
  "overlayPath": "/en/publications/wp/issues/2026/01/30/nowcasting-economic-growth-with-machine-learning-and-satellite-data-573623/index.md",
  "summary": "The absence of reliable data on fundamental economic indicators (e.g. real GDP), combined with structural shifts in the economy, can severely constrain the ability to conduct accurate macroeconomic analysis and forecasting.",
  "sections": [
    {
      "heading": "Overview",
      "content": "- Authors: Eurydice Fotopoulou, Iyke Maduako, M. Belen Sbrancia, Prachi Srivastava\n- Publication date: January 30, 2026\n- Series: Working Paper No. 2026/020\n- Core premise: Integrating machine learning and satellite data to estimate real GDP can address limitations from the absence of reliable data on fundamental economic indicators (e.g. real GDP) and structural shifts in the economy."
    },
    {
      "heading": "Key findings",
      "content": "- Incorporating satellite-based nightlight data into a random forest model significantly improves the accuracy of quarterly GDP growth estimates compared with models relying solely on traditional indicators.\n- The empirical application advances the nowcasting field to enhance economic forecasting in economies with significant data gaps."
    },
    {
      "heading": "Methodology and data",
      "content": "- Modeling approach: Random forest model\n- Novel data input: Satellite-based nightlight data\n- Comparison: Models with nightlight data versus models relying solely on traditional indicators\n- Analytical focus: Nowcasting quarterly GDP growth"
    },
    {
      "heading": "Applications and scope",
      "content": "- Geographic and topical relevance indicated by keywords and subjects: Caribbean, Central America, South America, Sub-Saharan Africa, Oil, Oil production, Consumption, Technology\n- Subject tags: Artificial intelligence (economics), Commodities, Econometric analysis, Economic and financial statistics, Economic forecasting, Nowcasting Economic Growth with Machine Learning and Satellite Data, Time series analysis"
    },
    {
      "heading": "Policy implications and contributions",
      "content": "- Offers an alternative approach for macroeconomic analysis and forecasting where reliable official statistics on real GDP are lacking or where structural shifts complicate standard methods.\n- Demonstrates the value of combining machine learning techniques with nontraditional data sources (satellite nightlights) for nowcasting and improving policy-relevant short-term GDP estimates.\n\n---\n\n Content in this bundle\n\n- Working Paper\n  - Working Paper (Markdown version){rel=\"alternate\" type=\"text/markdown\"}\n  - Working Paper (PDF){rel=\"external\" type=\"application/pdf\"}\n\n---\n\nSource: https://www.imf.org/en/publications/wp/issues/2026/01/30/nowcasting-economic-growth-with-machine-learning-and-satellite-data-573623"
    }
  ],
  "bullets": [
    "[Markdown version](/en/publications/wp/issues/2026/01/30/nowcasting-economic-growth-with-machine-learning-and-satellite-data-573623/index.md)",
    "[Structured JSON version](/en/publications/wp/issues/2026/01/30/nowcasting-economic-growth-with-machine-learning-and-satellite-data-573623/index.json)",
    "[Bundle manifest](/en/publications/wp/issues/2026/01/30/nowcasting-economic-growth-with-machine-learning-and-satellite-data-573623/bundle-manifest.json)",
    "Authors: Eurydice Fotopoulou, Iyke Maduako, M. Belen Sbrancia, Prachi Srivastava",
    "Published: January 30, 2026",
    "Series: IMF Working Papers",
    "DOI: https://doi.org/10.5089/9798229037471.001",
    "Authors: Eurydice Fotopoulou, Iyke Maduako, M. Belen Sbrancia, Prachi Srivastava",
    "Publication date: January 30, 2026",
    "Series: Working Paper No. 2026/020",
    "Core premise: Integrating machine learning and satellite data to estimate real GDP can address limitations from the absence of reliable data on fundamental economic indicators (e.g. real GDP) and structural shifts in the economy.",
    "Incorporating satellite-based nightlight data into a random forest model significantly improves the accuracy of quarterly GDP growth estimates compared with models relying solely on traditional indicators.",
    "The empirical application advances the nowcasting field to enhance economic forecasting in economies with significant data gaps.",
    "Modeling approach: Random forest model",
    "Novel data input: Satellite-based nightlight data",
    "Comparison: Models with nightlight data versus models relying solely on traditional indicators",
    "Analytical focus: Nowcasting quarterly GDP growth",
    "Geographic and topical relevance indicated by keywords and subjects: Caribbean, Central America, South America, Sub-Saharan Africa, Oil, Oil production, Consumption, Technology",
    "Subject tags: Artificial intelligence (economics), Commodities, Econometric analysis, Economic and financial statistics, Economic forecasting, Nowcasting Economic Growth with Machine Learning and Satellite Data, Time series analysis",
    "Offers an alternative approach for macroeconomic analysis and forecasting where reliable official statistics on real GDP are lacking or where structural shifts complicate standard methods.",
    "Demonstrates the value of combining machine learning techniques with nontraditional data sources (satellite nightlights) for nowcasting and improving policy-relevant short-term GDP estimates.",
    "**Working Paper**"
  ],
  "related": [
    {
      "title": "Working Paper",
      "role": "paper",
      "sourceUrl": "https://www.imf.org/-/media/files/publications/wp/2026/english/wpiea2026020-source-pdf.pdf",
      "summary": {
        "path": "/-/media/files/publications/wp/2026/english/wpiea2026020-source-pdf.pdf.md",
        "mime": "text/markdown"
      },
      "binary": {
        "path": "/-/media/files/publications/wp/2026/english/wpiea2026020-source-pdf.pdf",
        "mime": "application/pdf"
      }
    }
  ],
  "alternates": {
    "markdown": "/en/publications/wp/issues/2026/01/30/nowcasting-economic-growth-with-machine-learning-and-satellite-data-573623/index.md",
    "json": "/en/publications/wp/issues/2026/01/30/nowcasting-economic-growth-with-machine-learning-and-satellite-data-573623/index.json",
    "bundleManifest": "/en/publications/wp/issues/2026/01/30/nowcasting-economic-growth-with-machine-learning-and-satellite-data-573623/bundle-manifest.json"
  },
  "generatedAtUtc": "2026-09-17T02:55:17.024Z"
}