Searcher

Published on Thursday, September 17, 2020

Big Data techniques used

Global | Investment in Real Time and High Definition

The high uncertainty triggered by the Covid-19 crisis has stressed the need to monitor the evolution of the economy in “real time”. We developed a set of Big Data indicators to track Investment in Real Time and High Definition, extending the analysis to assess the business cycle and complementing our tracker of consumption.

Key points

  • Key points:
  • Economists normally rely on information by Statistical Agencies and Central Banks (hard data) and private surveys to firms from the manufacturing and services sectors (soft data) to assess the state of the business cycle. While the official information is carefully designed to understand the state of the business cycle in terms of consistency it also has some shortcomings such as some time lags in the official releases and the qualitative information included in the surveys.
  • A new strand of the empirical literature has focused on tracking the economic activity in real time and high definition, particularly since the Covid-19, given the need to assess the cycle after this crisis. Analysts and Central Banks have reacted either by focusing on more timely soft data indicators like surveys or developing high frequency weekly models (such as the new Weekly economic indexes released by the FED & Bundesbank), while some researchers have started to develop new indicators building on Big Data information.
  • Following our previous work tracking consumption with credit and debit cards transactions, we introduce now a set of novel Big Data indicators to track Investment in Real Time and High Definition extending the individual-to-firm credit/debit card transactions with firm-to-firm bank transactions. We develop a new Big Data investment indicator in real time, with sectorial and geographical disaggregation for Turkey using BBVA transaction data. The results are successful and validated by cross correlations with the official figures, with alternative higher frequency proxies of investment and in terms of improving the forecasting accuracy of our standard nowcasting Dynamic Factor Models.
  • These positive results could be extended to other emerging countries like Mexico and Colombia, but also to developed countries such as Spain, where we also find high correlations of the estimated indices with official figures and monthly proxies of investment. While Turkey´s investment is recovering relatively faster in construction and machinery investment, Spain, Mexico and Colombia show a weaker recovery with construction bottoming out but still lagging machinery investment.

Documents to download

  • Report (PDF)

    Big-Data-EW_-Investment-in-Real-Time-and-High-Definition_September-2020.pdf English September 17, 2020

Geographies

Topics

New comment

Be the first to add a comment.

Load more

You may also be interested in