MICA-BBVA: a factor model of economic and financial indicators for short-term GDP forecasting

Abstract

In this paper we extend the Stock and Watson's (Leading economic indicators, new approaches and forecasting records, 1991) single-index dynamic factor model in an econometric framework that has the advantage of combining information from real and financial indicators published at different frequencies and delays with respect to the period to which they refer. We find that the common factor reflects the behavior of the Spanish business cycle well. We also show that financial indicators are useful for forecasting output growth, particularly when certain financial variables lead the common factor. Finally, we provide a simulated real-time exercise and prove that the model is a very useful tool for the short-term analysis of the Spanish Economy.

Robust Implementation of a Parsimonious Dynamic Factor Model to Nowcast GDP

“…States, Camacho and Doménech (2012) for Spain, Angelini et al (2008a), Angelini et al (2008b), Camacho and Perez-Quiros (2010), and Camacho and Garcia-Serrador (2013) for the Euro Area. 2 They modify the "exact" DFM by Stock and Watson (1991) to account for problems of different frequency and asynchronous publication of series underlying the real-time forecast in applying a Kalman Filter strategy to fill up the series.…”

“…Our sample application consists in nowcasting Spanish GDP growth at a monthly frequency, which through official channels is available only in quarterly frequency. Our selection of indicators from a "universe of potentially available time series" (Camacho and Doménech, 2012) proceeds in two steps. First, we reduce a set of 258 time series related to Spanish GDP and available at monthly frequency to 27 series by keeping only those series showing a reasonable correlation with the series to be nowcasted both contemporaneously and at its quarterly and yearly lags.…”

“…However, for industrialized economies in the contemporary world GDP data is published quarterly and information on GDP growth usually becomes available with a lag of six weeks seen from the end of a particular quarter through so-called flash or first releases. 1 First releases are early announcements of second releases also referred to as final estimates (Camacho and Perez-Quiros, 2010;Camacho and Doménech, 2012). The latter are released about 14 weeks after the respective quarter has ended and are possibly subject to revisions and corrections.…”

A chronology of turning points in economic activity: Spain, 1850–2011

“…We begin with the early methods that Gerhard Bry and Charlotte Boschan introduced in 1971 at the NBER. The Bry and Boschan (1971) algorithm comes closest to translating the NBER's definition into practice: remove seasonals from the data in levels (there is no need to detrend), smooth the data lightly to remove aberrations, then identify local minima and maxima in the series. The local minima and maxima are constrained to ensure that cycles have a certain duration, that they alternate, and that complete cycles last at least 15 months.…”

“…A local minimum is a trough and the following local maximum a peak, so that the period between trough and peak is an expansion, and from peak to trough a recession. The Bry and Boschan (1971) algorithm was applied to quarterly data by Harding and Pagan (2002a,b), Kose et al (2003Kose et al ( , 2008a, among others. As arbitrary as the Bry and Boschan (1971) algorithm may seem, it is simple to implement, reproducible, and perhaps more critically, it does not require that the data be detrended.…”