Real-Time Forecasting of GDP Based on a Large Factor Model with Monthly and Quarterly Data

By Christian Schumacher, J. Breitung. Social Science Research Network, 2008

262 citations recorded by the source corpus.

This paper discusses a factor model for estimating monthly GDP using a large number of monthly and quarterly time series in real-time. To take into account the different periodicities of the data and missing observations at the end of the sample, the factors are estimated by applying an EM algorithm combined with a principal components estimator. We discuss the in-sample properties of the estimator in real-time environments and methods for out-of-sample forecasting. As an empirical application, we estimate monthly German GDP in real-time, discuss the nowcast and forecast accuracy of the model and the role of revisions. Furthermore, we assess the contribution of timely monthly data to the forecast performance.

Subfields: Economics and Econometrics

View Paper: https://www.semanticscholar.org/paper/48a60528782d1258c2c190b2c11c19a4e0079476

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

Citations in the corpus

Macroeconomic Forecasting With Mixed-Frequency Data (2008, methodology): “There also are factor model approaches that make use of mixed-frequency data, such as the model of Schumacher and Breitung (2006), which adapt single-frequency factor models (see Boivin and Ng 2005 for a review).”
A Survey of Econometric Methods for Mixed-Frequency Data (2013, unknown): “…components in the update: the common component from the previous iteration F̂ (j−1)Λ̂(j−1)i , and a low-frequency idiosyncratic component Xobsi − AiF̂ (j−1)Λ̂ (j−1) i , distributed by the projection coeffi cient A ′ i (A ′ iAi) −1 on the high-frequency periods, see Breitung and Schumacher (2008).”
A Look into the Factor Model Black Box: Publication Lags and the Role of Hard and Soft Data in Forecasting GDP (2007, unknown): “In relating the individual indicators to a few latent factors, weights are implicitly attached to the former in a parsimonious way (e.g. Stock and Watson, 2002; Forni et al., 2003; Breitung and Schumacher, 2006; Altissimo et al. 2007).”
On factor models with random missing: EM estimation, inference, and cross validation (2020, methodology): “Schumacher and Breitung (2008) apply Stock and Watson’s methodology to nowcast German gross domestic product (GDP).”
State Space Models and MIDAS Regressions (2013, methodology): “…data (see e.g. Armesto, Hernandez-Murillo, Owyang, and Piger (2008), Clements and Galvão (2008a), Clements and Galvão (2008b), Galvão (2006), Schumacher and Breitung (2008), Tay (2007)), or improving quarterly and monthly macroeconomic predictions with daily financial data (see e.g.…”
Estimating and Forecasting the Euro Area Monthly National Accounts from a Dynamic Factor Model (2010, unknown): “by Mariano and Murasawa (2003) or Breitung and Schumacher (2006).”
Maximum Likelihood Estimation of Factor Models on Data Sets with Arbitrary Pattern of Missing Data (2010, methodology): “The latter is a popular method based on EM to calculate principal components from a panel with missing data, see e.g. Schumacher and Breitung (2008).”
Can Google Data Help Predict French Youth Unemployment (2012, background): “We also thank the referee errors or omissions. sité du Maine, Avenue Olivier Karamé). rights reserved. there is a growing literature on nowcasting, i.e., predicting the present (see, for instance, Castle et al., 2009; Giannone et al., 2008; Schumacher and Breitung, 2008, among many others).”