New Eurocoin: Tracking Economic Growth in Real Time

Abstract

Removal of short-run dynamics from a stationary time series to isolate the medium- to long-run component can be obtained by a bandpass filter. However, bandpass filters are infinite moving averages and can therefore deteriorate at the end of the sample. This is a well-known result in the literature isolating the business cycle in integrated series. We show that the same problem arises with our application to stationary time series. In this paper, we develop a method to obtain smoothing of a stationary time series by using only contemporaneous values of a large data set, so that no end-of-sample deterioration occurs. Our method is applied to the construction of New Eurocoin, an indicator of economic activity for the euro area, which is an estimate, in real time, of the medium- to long-run component of GDP growth. As our data set is monthly and most of the series are updated with a short delay, we are able to produce a monthly real-time indicator. As an estimate of the medium- to long-run GDP growth, Eurocoin performs better than the bandpass filter at the end of the sample in terms of both fitting and turning-point signaling. (c) 2010 The President and Fellows of Harvard College and the Massachusetts Institute of Technology.

Short term forecasting of national growth

“…This is very relevant for economic policy. Altissimo et al (2010) show that the same problem arises with application to stationary time series. And, through New Eurocoin, they develop a method to obtain smoothing of a stationary time series so as to avoid the occurrence of end-of-sample deterioration and short-run fluctuations.…”

“…Estimations of NE are obtained through the generalized dynamic factor model. The specific references are Altissimo et al (2001Altissimo et al ( , 2010, Forni et al ( , 2004Forni et al ( , 2005. The main empirical contribution of this research is to propose some procedures to estimate the following disaggregated Antonio Frenda : frendaantonio@libero.it A.…”

Small- Versus Big-Data Factor Extraction in Dynamic Factor Models: An Empirical Assessment

“…Furthermore, as it is usual in this literature, they have been stan-dardized to have zero mean and variance one. The variables can be classi…ed into the following 12 economic categories: industrial production (13); employment and unemployment (44); housing starts (8); inventories, orders and sales (8); prices (16); earnings (3); interest rates and spreads (18); money and credit (14); stock prices (3); housing prices (2); exchange rates (6); and other (2), with the number of the series in the category given in parentheses. In order to obtain a balanced panel, we select those variables observed without missing values over the whole observation period.…”

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

“…The set of monthly indicators used to compute the forecasts comprises the monthly growth rates of industrial production, employment, income and sales, which become available in di¤erent time periods and with di¤erent publication delays. 1 At any point t in real time, we simply use the time-t data vintage to extract the short-term forecast of the next unobserved US GDP …gure. As time progresses, we re-estimate all the models for each period, always using the latest data vintage to compute the forecast.…”

“…In section 3, we review the main models used for this purpose address the role of the number of series in factor models. 1 Although the NBER Business Cycle Dating Committee does not have a …xed de…nition of economic activity, it acknowledges on its home page that these are the …ve indicators that examines to analyze the US business cycle conditions. 2 After the paper was reviewed by the editor, we became aware of the working paper by Foroni and Marcellino (2013).…”