Short-Term Forecasting for Empirical Economists. A Survey of the Recently Proposed Algorithms

Abstract

Practitioners do not always use research …ndings, sometimes because the research is not always conducted in a manner relevant to real-world practice. This survey seeks to close the gap between research and practice on short-term forecasting in real time.Towards this end, we review the most relevant recent contributions to the literature, examine their pros and cons, and we take the liberty of proposing some lines of future research. We include bridge equations, MIDAS, VARs, factor models and Markovswitching factor models, all allowing for mixed-frequency and ragged ends. Using the four constituent monthly series of the Stock-Watson coincident index, industrial production, employment, income and sales, we evaluate their empirical performance to forecast quarterly US GDP growth rates in real time. Finally, we review the main results regarding the number of predictors in factor based forecasts and how the selection of the more informative or representative variables can be made.Keywords: Business Cycles, Output Growth, Time Series. JEL Classi…cation: E32, C22, E27C o rre sp o n d in g A u th o r: G a b rie l P e re z Q u iro s, B a n c o d e E sp a ñ a , R e se a rch D iv isio n , M o n e ta ry a n d F in a n c ia l S tu d ie s D e p a rtm e nt, D G E c o n o m ic s, S ta tistic s, a n d R e se a rch . A lc a lá 4 8 , 2 8 0 1 4 M a d rid , S p a in . E -m a il: g a b rie l.p e re z @ b d e .e s. M . C a m a ch o a n d P. P o n c e la a ck n ow le d g e s su p p o rt fro m p ro je c ts E C O 2 0 1 0 -1 9 8 3 0 , E C O 2 0 0 9 -1 0 2 8 7 a n d E C O 2 0 1 2 -3 2 8 5 4 , re sp e c tive ly.

Google data in bridge equation models for German GDP

“…Thus, intuitively speaking, the information contained in various short-term indicators gets transferred, or "bridged", to the coherent structure implied by the National Accounts (Wohlrabe, 2008). There exist many applications of BEM's in the literature, among which are Angelini et al (2011), Baffigi et al (2004), Camacho et al (2013), Foroni and Marcellino (2014) and Schumacher (2014), the latter comparing MI(xed) DA(ta) S(ampling) (Ghysels et al, 2004) and BEM's as compet-itive approaches to dealing with the mixed-frequency characteristic of many (macro)economic datasets.…”

Should One Follow Movements in the Oil Price or in Money Supply? Forecasting Quarterly GDP Growth in Russia with Higher-Frequency Indicators

“…They are easy to implement and transparent. Literature applications include Ingenito and Trehan (1996), Baffigi et al (2004), Golinelli and Parigi (2007), Diron (2008), Hahn and Skudelny (2008), Rünstler et al (2009), Bulligan et al (2010), Angelini et al (2011), Camacho et al (2013), , Schumacher (2014) and Bulligan et al (2015). This paper employs the classical bridge equations procedure as outlined in Schumacher (2016).…”

With or Without You - Do Financial Data Help to Forecast Industrial Production?

“…1 Banbura et al (2011), Banbura et al (2013 or Camacho et al (2013) provide some complementary overviews of the existing literature on short-term forecasting.…”