{
  "title": "GDP Nowcasting Performance of Traditional Econometric Models vs Machine-Learning Algorithms: Simulation and Case Studies",
  "publication": "IMF Working Papers, December 5, 2025",
  "sourceUrl": "https://www.imf.org/en/publications/wp/issues/2025/12/05/gdp-nowcasting-performance-of-traditional-econometric-models-vs-machine-learning-572360",
  "canonical": "https://www.imf.org/en/publications/wp/issues/2025/12/05/gdp-nowcasting-performance-of-traditional-econometric-models-vs-machine-learning-572360",
  "overlayPath": "/en/publications/wp/issues/2025/12/05/gdp-nowcasting-performance-of-traditional-econometric-models-vs-machine-learning-572360/index.md",
  "summary": "Are Machine Learning (ML) algorithms superior to traditional econometric models for GDP nowcasting in a time series setting?",
  "sections": [
    {
      "heading": "Summary findings",
      "content": "- Traditional econometric models tend to outperform Machine Learning (ML) algorithms for GDP nowcasting in a time series setting.\n- Among ML algorithms, linear ML algorithms—Lasso and Elastic Net—perform best in nowcasting and can surpass traditional econometric models in cases of long GDP data and rich high-frequency indicators.\n- Among traditional econometric models, the Bridge and Dynamic Factor models deliver the strongest empirical results.\n- Three-Pass Regression Filter performs well in the authors’ simulation.\n- Complex and non-linear ML algorithms are prone to overfitting due to the relatively short length of GDP series, compromising their out-of-sample performance."
    },
    {
      "heading": "Performance by model class",
      "content": "- Traditional econometric models\n  - Tend to outperform ML algorithms in nowcasting across simulation and six country cases.\n  - Bridge model and Dynamic Factor model show strongest empirical results.\n  - Three-Pass Regression Filter shows strong performance in simulation settings.\n- Machine-Learning algorithms\n  - Linear ML algorithms (Lasso and Elastic Net) provide the best ML performance and can outperform traditional models when data conditions include long GDP series and rich high-frequency indicators.\n  - Complex and non-linear ML algorithms generally underperform out-of-sample due to overfitting driven by short GDP time series."
    },
    {
      "heading": "Key empirical scope and evidence",
      "content": "- Evaluation covers all models from both classes ever used in nowcasting across simulation and six country cases.\n- Performance conclusions are drawn from both simulation results and empirical case studies."
    },
    {
      "heading": "Reasons, limitations, and methodological considerations",
      "content": "- Short length of GDP series:\n  - Causes complex and non-linear ML algorithms to be prone to overfitting.\n  - Overfitting compromises out-of-sample performance for non-linear ML methods.\n- Data richness and series length:\n  - Linear ML methods (Lasso, Elastic Net) can outperform traditional econometric models when GDP series are long and high-frequency indicators are abundant."
    },
    {
      "heading": "Implications for practitioners and researchers",
      "content": "- For GDP nowcasting in typical time series contexts (short GDP series), prefer traditional econometric models—particularly Bridge and Dynamic Factor models.\n- Consider linear ML methods (Lasso, Elastic Net) when:\n  - GDP data are long, and\n  - Rich high-frequency indicators are available.\n- Exercise caution with complex and non-linear ML algorithms due to overfitting risk in short time series settings; emphasize out-of-sample validation and parsimonious modeling.\n\nIMF Working Paper by Klakow Akepanidtaworn and Korkrid Akepanidtaworn, December 5, 2025.\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/2025/12/05/gdp-nowcasting-performance-of-traditional-econometric-models-vs-machine-learning-572360"
    }
  ],
  "bullets": [
    "[Markdown version](/en/publications/wp/issues/2025/12/05/gdp-nowcasting-performance-of-traditional-econometric-models-vs-machine-learning-572360/index.md)",
    "[Structured JSON version](/en/publications/wp/issues/2025/12/05/gdp-nowcasting-performance-of-traditional-econometric-models-vs-machine-learning-572360/index.json)",
    "[Bundle manifest](/en/publications/wp/issues/2025/12/05/gdp-nowcasting-performance-of-traditional-econometric-models-vs-machine-learning-572360/bundle-manifest.json)",
    "Authors: Klakow Akepanidtaworn, Korkrid Akepanidtaworn",
    "Published: December 5, 2025",
    "Series: IMF Working Papers",
    "DOI: https://doi.org/10.5089/9798229033626.001",
    "Traditional econometric models tend to outperform Machine Learning (ML) algorithms for GDP nowcasting in a time series setting.",
    "Among ML algorithms, linear ML algorithms—Lasso and Elastic Net—perform best in nowcasting and can surpass traditional econometric models in cases of long GDP data and rich high-frequency indicators.",
    "Among traditional econometric models, the Bridge and Dynamic Factor models deliver the strongest empirical results.",
    "Three-Pass Regression Filter performs well in the authors’ simulation.",
    "Complex and non-linear ML algorithms are prone to overfitting due to the relatively short length of GDP series, compromising their out-of-sample performance.",
    "Traditional econometric models",
    "Machine-Learning algorithms",
    "Evaluation covers all models from both classes ever used in nowcasting across simulation and six country cases.",
    "Performance conclusions are drawn from both simulation results and empirical case studies.",
    "Short length of GDP series:",
    "Data richness and series length:",
    "For GDP nowcasting in typical time series contexts (short GDP series), prefer traditional econometric models—particularly Bridge and Dynamic Factor models.",
    "Consider linear ML methods (Lasso, Elastic Net) when:",
    "Exercise caution with complex and non-linear ML algorithms due to overfitting risk in short time series settings; emphasize out-of-sample validation and parsimonious modeling.",
    "**Working Paper**"
  ],
  "related": [
    {
      "title": "Working Paper",
      "role": "paper",
      "sourceUrl": "https://www.imf.org/-/media/files/publications/wp/2025/english/wpiea2025252-source-pdf.pdf",
      "summary": {
        "path": "/-/media/files/publications/wp/2025/english/wpiea2025252-source-pdf.pdf.md",
        "mime": "text/markdown"
      },
      "binary": {
        "path": "/-/media/files/publications/wp/2025/english/wpiea2025252-source-pdf.pdf",
        "mime": "application/pdf"
      }
    }
  ],
  "alternates": {
    "markdown": "/en/publications/wp/issues/2025/12/05/gdp-nowcasting-performance-of-traditional-econometric-models-vs-machine-learning-572360/index.md",
    "json": "/en/publications/wp/issues/2025/12/05/gdp-nowcasting-performance-of-traditional-econometric-models-vs-machine-learning-572360/index.json",
    "bundleManifest": "/en/publications/wp/issues/2025/12/05/gdp-nowcasting-performance-of-traditional-econometric-models-vs-machine-learning-572360/bundle-manifest.json"
  },
  "generatedAtUtc": "2026-09-17T02:42:53.137Z"
}