Etla currently has two projects that use big data in economic forecasting.
ETLAnow unemployment predicts the unemployment rate in every EU country using real-time data from Google Trends database and the latest official data from Eurostat. Forecasts three months ahead are automatically updated every morning, and are also available on Twitter. ETLAnow is a collaborative work project involving 29 European research institutes.
ETLAnow is discussed in detail in several documents. ”ETLAnow: A Model for Forecasting with Big Data” provides an overview of how the model works. A more extensive analysis on unemployment forecasting using Google data is provided in the working paper ”Forecasting Unemployment with Google Searches” using U.S. data. Earlier work also includes this brief, which is based on our first trial for real-time forecasting and an earlier report in Finnish:”Big Data: Google-haut ennustavat työttömyyttä Suomessa”.
ETLAnow housing (in Finnish) predicts the housing prices in Finland using real-time data from the Google Trends database and the latest official data from Statistics Finland. The Google searches used in the forecast are related to demand for housing. Forecasts three months ahead are automatically updated every morning.
The ETLAnow housing project is documented in Etla Report Number 63 (in Finnish): ”Google-haut Suomen asuntojen hintojen ennustajana”. The model is quite similar to the model used in ETLAnow unemployment. The housing project is currently on -hold.