Euro area GDP forecasting using large survey datasets A random forest approach. Biau, O. & D 'elia, A. 2015.
Euro area GDP forecasting using large survey datasets A random forest approach [pdf]Website  abstract   bibtex   
Recent works in the econometric literature consider the problem of efficiently summarising a large set of variables and using this summary for a variety of purposes, including forecasts (Stock and Watson, 2002; Forni et al., 2005; Giannone et al., 2008; for a wide review, see Eklund and Kapetanios, 2008). Factor analysis combined with linear modelling has usually been the main tool used for this task. This paper presents a new statistical approach to forecasting macro-economic aggregates, based on the Random Forests technique, originally developed as a learning classification tool (Breiman, 2001). This technique can handle a very large number of input variables without overfitting and is known to enjoy good prediction properties and to be robust to noise.

Downloads: 0