{
  "title": "An Algorithmic Crystal Ball: Forecasts-based on Machine Learning",
  "publication": "IMF Working Papers, November 1, 2018",
  "sourceUrl": "https://www.imf.org/en/publications/wp/issues/2018/11/01/an-algorithmic-crystal-ball-forecasts-based-on-machine-learning-46288",
  "canonical": "https://www.imf.org/en/publications/wp/issues/2018/11/01/an-algorithmic-crystal-ball-forecasts-based-on-machine-learning-46288",
  "overlayPath": "/en/publications/wp/issues/2018/11/01/an-algorithmic-crystal-ball-forecasts-based-on-machine-learning-46288/index.md",
  "summary": "Forecasting macroeconomic variables is key to developing a view on a country's economic outlook. Most traditional forecasting models rely on fitting data to a pre-specified relationship between input and output variables, thereby assuming a specific functional and stochastic process underlying that",
  "sections": [
    {
      "heading": "Research overview",
      "content": "- Forecasting macroeconomic variables is key to developing a view on a country's economic outlook.\n- Traditional forecasting models rely on fitting data to a pre-specified relationship between input and output variables, assuming a specific functional and stochastic process.\n- The paper pursues a data-driven approach using machine learning algorithms that impose limited restrictions on the nature of the true relationship between input and output variables.\n- Algorithms applied: Elastic Net, SuperLearner, and Recurring Neural Network.\n- Empirical application: macro data for seven, broadly representative, advanced and emerging economies."
    },
    {
      "heading": "Main findings",
      "content": "- Machine learning algorithms can outperform traditional statistical models in forecasting macroeconomic variables.\n- These algorithms offer a relevant addition to the field of economic forecasting by improving predictive performance relative to conventional approaches.\n- The study specifically evaluates performance against benchmark WEO performance and WEO forecast comparisons (keywords indicate benchmarking to WEO)."
    },
    {
      "heading": "Methodology and data notes",
      "content": "- Algorithms implemented: Elastic Net, SuperLearner, Recurring Neural Network.\n- Data context: time series, training data, test data, macro variables including real GDP and GDP growth nowcasts.\n- Comparative approaches referenced: decision tree algorithm, Random Forest algorithm, Random Forest algorithms to nowcast GDP growth.\n- Emphasis on limited structural assumptions about the generation process of forecasts."
    },
    {
      "heading": "Implications and relevance",
      "content": "- Machine learning provides a complementary toolkit for macroeconomic forecasting tasks, particularly for nowcasting and forecasting real GDP.\n- Results support exploring machine learning models alongside traditional econometric models for national economic outlook assessments.\n\n---\n\n Content in this bundle\n\n- An Algorithmic Crystal Ball: Forecasts-based on Machine Learning, WP/18/230, November 2018\n  - An Algorithmic Crystal Ball: Forecasts-based on Machine Learning, WP/18/230, November 2018 (Markdown version){rel=\"alternate\" type=\"text/markdown\"}\n  - An Algorithmic Crystal Ball: Forecasts-based on Machine Learning, WP/18/230, November 2018 (PDF){rel=\"external\" type=\"application/pdf\"}\n\n---\n\nSource: https://www.imf.org/en/publications/wp/issues/2018/11/01/an-algorithmic-crystal-ball-forecasts-based-on-machine-learning-46288"
    }
  ],
  "bullets": [
    "[Markdown version](/en/publications/wp/issues/2018/11/01/an-algorithmic-crystal-ball-forecasts-based-on-machine-learning-46288/index.md)",
    "[Structured JSON version](/en/publications/wp/issues/2018/11/01/an-algorithmic-crystal-ball-forecasts-based-on-machine-learning-46288/index.json)",
    "[Bundle manifest](/en/publications/wp/issues/2018/11/01/an-algorithmic-crystal-ball-forecasts-based-on-machine-learning-46288/bundle-manifest.json)",
    "Authors: Jin-Kyu Jung, Manasa Patnam, Anna Ter-Martirosyan",
    "Published: November 1, 2018",
    "Series: IMF Working Papers",
    "DOI: https://doi.org/10.5089/9781484380635.001",
    "Forecasting macroeconomic variables is key to developing a view on a country's economic outlook.",
    "Traditional forecasting models rely on fitting data to a pre-specified relationship between input and output variables, assuming a specific functional and stochastic process.",
    "The paper pursues a data-driven approach using machine learning algorithms that impose limited restrictions on the nature of the true relationship between input and output variables.",
    "Algorithms applied: Elastic Net, SuperLearner, and Recurring Neural Network.",
    "Empirical application: macro data for seven, broadly representative, advanced and emerging economies.",
    "Machine learning algorithms can outperform traditional statistical models in forecasting macroeconomic variables.",
    "These algorithms offer a relevant addition to the field of economic forecasting by improving predictive performance relative to conventional approaches.",
    "The study specifically evaluates performance against benchmark WEO performance and WEO forecast comparisons (keywords indicate benchmarking to WEO).",
    "Algorithms implemented: Elastic Net, SuperLearner, Recurring Neural Network.",
    "Data context: time series, training data, test data, macro variables including real GDP and GDP growth nowcasts.",
    "Comparative approaches referenced: decision tree algorithm, Random Forest algorithm, Random Forest algorithms to nowcast GDP growth.",
    "Emphasis on limited structural assumptions about the generation process of forecasts.",
    "Machine learning provides a complementary toolkit for macroeconomic forecasting tasks, particularly for nowcasting and forecasting real GDP.",
    "Results support exploring machine learning models alongside traditional econometric models for national economic outlook assessments.",
    "**An Algorithmic Crystal Ball: Forecasts-based on Machine Learning, WP/18/230, November 2018**"
  ],
  "related": [
    {
      "title": "An Algorithmic Crystal Ball: Forecasts-based on Machine Learning, WP/18/230, November 2018",
      "role": "document",
      "sourceUrl": "https://www.imf.org/-/media/files/publications/wp/2018/wp18230.pdf",
      "summary": {
        "path": "/-/media/files/publications/wp/2018/wp18230.pdf.md",
        "mime": "text/markdown"
      },
      "binary": {
        "path": "/-/media/files/publications/wp/2018/wp18230.pdf",
        "mime": "application/pdf"
      }
    }
  ],
  "alternates": {
    "markdown": "/en/publications/wp/issues/2018/11/01/an-algorithmic-crystal-ball-forecasts-based-on-machine-learning-46288/index.md",
    "json": "/en/publications/wp/issues/2018/11/01/an-algorithmic-crystal-ball-forecasts-based-on-machine-learning-46288/index.json",
    "bundleManifest": "/en/publications/wp/issues/2018/11/01/an-algorithmic-crystal-ball-forecasts-based-on-machine-learning-46288/bundle-manifest.json"
  },
  "generatedAtUtc": "2026-09-18T14:15:17.434Z"
}