Published on Wednesday, October 14, 2020

Document number 20/13

Global | Investment in Real Time and High Definition: A Big Data Approach

We introduce a set of Big Data indicators to mimic Investment in Real Time and High Definition. We use aggregate information of BBVA Big Data from firm-to-firm transactions and complement with individual-to-firm information. The results are successful and are validated through multiple robustness checks.

Key points

  • Key points:
  • The high granularity and geo-localization of individual and firm-to-firm transactions data allow us to track the investment data in "high definition" from both a sectoral and geographical perspective.
  • Particularly, we replicate the quarterly national accounts aggregate investment (gross fixed capital formation) and its bigger components (Machinery & Equipment and Construction) in real time for the case of Turkey.
  • The inclusion of the development indicators improve the properties of a standard Dynamic Factor Model for Nowcasting including forecasting accuracy, anticipation and news contributions.
  • We also extend part of our tests to other countries, such as Spain and Mexico, with positive results.

Documents to download

  • Report (PDF)

    Working_Paper_Investment_in_Real_Time_and_High_Definition_A_Big_Data_Approach_WB.pdf English October 14, 2020



New comment

Be the first to add a comment.

Load more

You may also be interested in