A note on the predictive power of survey data in nowcasting euro area GDP

AuthorJeong‐Ryeol Kurz‐Kim
Published date01 September 2019
DOIhttp://doi.org/10.1002/for.2578
Date01 September 2019
Received: 23 April 2018 Revised: 5 December 2018 Accepted: 31 January 2019
DOI: 10.1002/for.2578
RESEARCH ARTICLE
A note on the predictive power of survey data in nowcasting
euro area GDP
Jeong-Ryeol Kurz-Kim
Deutsche Bundesbank, Frankfurt am
Main, Germany
Correspondence
Jeong-Ryeol Kurz-Kim, Deutsche
Bundesbank, Wilhelm-Epstein-Strasse 14,
60431 Frankfurt am Main, Germany.
Email:
jeong-ryeol.kurz-kim@bundesbank.de
Abstract
This paper investigates the trade-off between timeliness and quality in now-
casting practices. This trade-off arises when the frequency of the variable to
be nowcast, such as gross domestic product (GDP), is quarterly, while that of
the underlying panel data is monthly; and the latter contains both survey and
macroeconomic data. These two categories of data have different properties
regarding timeliness and quality: the survey data are timely available(but might
possess less predictive power), while the macroeconomic data possess more pre-
dictive power (but are not timely available because of their publication lags). In
our empirical analysis, we use a modified dynamic factor model which takes
three refinements for the standard dynamic factor model of Stock and Watson
(Journal of Business and Economic Statistics, 2002, 20, 147–162) into account,
namely mixed frequency, preselections and cointegration among the economic
variables. Our main finding from a historical nowcasting simulation based on
euro area GDP is that the predictive power of the survey data depends on the
economic circumstances; namely, that survey data are more useful in tranquil
times, and less so in times of turmoil.
KEYWORDS
co-integration, dynamic factor model, mixed frequency, nowcasting, preselections, survey data
1INTRODUCTION
The current-quarter forecasting of gross domestic product
(GDP)—usually called nowcasting—is a useful guide to
understanding the current state of economic activity. For
this purpose, a raft of indicators is often used. Forecasters
usually divide these indicators into two groups: soft and
hard indicators. A typical example of soft indicators is sur-
vey data, while macroeconomic data are a typical example
of hard indicators. The reason for making this division
is the trade-off between timeliness and the quality of the
indicators: survey data have (almost) no time lag for the
corresponding month of the reference quarter, but they
are not part of the GDP calculation. Conversely, macroe-
conomic data, such as industrial production, are part of
GDP and hence possess a higher quality in prediction for
GDP, but they are published with some time lags. Other
macroeconomic data, such as unemployment rates and/or
car registrations, are not directly part of GDP, but they still
have a high correlation with GDP. Therefore, for empirical
nowcasting practices, it is useful to know how the trade-off
between timeliness and quality works.
The empirical consensus on this issue is that survey
data have useful indicators for GDP nowcasting, but their
relevance weakens when hard data areavailable; see Gian-
none, Reichlin, and Small (2008), for example. Specifically,
Girardi, Gayer, and Reuter (2015) set a hypothetical sce-
nario (where both groups of data are available without any
Journal of Forecasting. 2019;38:489–503. wileyonlinelibrary.com/journal/for © 2019 John Wiley & Sons, Ltd. 489

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT