Close panel

  • Home
  • Publications
  • Who we are
  • Big Data
  • Forecasts
    Searcher

    Published on Thursday, June 17, 2021

    Big Data techniques used

    Colombia | BBVA GDP Tracker: a new model for real time GDP forecast

    Summary

    A model, based on Camacho and Domenech (2012) and Aruoba and Diebold (2010), which are extensions of Stock and Watson (1991), was built to forecast Colombia's GDP in real time. Going forward, BBVA Research will continue to publish updates of the monthly results of this work.

    Key points

    • Key points:
    • Nine economic indicators and two big data indicators (consumption and investment) were selected to build a real-time projection model for GDP.
    • The model fit is good, although forecast accuracy one period ahead was lower when large drops occurred as a result of covid-19.
    • When there is a sharp change in the GDP trend, big data indicators improve the forecast considerably.
    • The projection for the economic monitoring index (ISE) for April 2021 is between 15.5% and 16.2%. For May 2021, the margin is wider, between 7.7% and 11.6%.

    Geographies

    Authors

    María Paula Castañeda
    Mauricio Hernández BBVA Research - Principal Economist
    Máximo Camacho Murcia University - External partner
    Agustín García BBVA Research - Lead Economist

    Documents and files

    Presentation (PDF)

    Informe-final-maqueteado.pdf

    Spanish - June 17, 2021

    New comment

    Be the first to add a comment.

    Load more

    You may also be interested in