Econometric forecasting of imports and exports indexes in Romania

AuthorSimionescu, M. - Bilan, Y.
PositionInstitute for Economic Forecasting, The Romanian Academy - National University of Food Technologies, Kyiv, Ukraine
Pages107-112
Bulletin of the Transilvania University of Braşov
Series V: Economic Sciences Vol. 6 (55) No. 2 - 2013
ECONOMETRIC FORECASTING OF
IMPORTS AND EXPORTS INDEXES
IN ROMANIA
M. SIMIONESCU
1 Y. BILAN2
Abstract: The objective of this research is the assessment of a ccuracy for
imports a nd exports predictions based on econometr ic models. For the
Romanian economy, the indexes of exports and imports a re forecasted on the
horizon 2011-2013. F or the first period (2011-2012), all the forecasts ar e
overestimated, this being a n important clue tha t these predictions ba sed on
econometric models did not ta ke into account the shocks in the economy. The
imports’ indexes anticipations are more accurate than those made for the
exports’ indexes. Moreover, the ex-a nte evalu ation of the predictions was
made for 2013, under the assumption that this year the indica tors would have
the value from 2012. An underestima tion of the expor ts is expected for 2013.
Key words: foreca sts, accuracy, econometric models, exports, imports.
1. Introduction
There are many quantitative methods
used in forecasting imports and exports, the
econometric models being the most used
tool. The relationships between
macroeconomic variables permanently
affected by changes, mostly in unstable
periods like crisis times, are better put into
evidence using econometric modelling. The
description of the macroeconomic variables
evolution is not enough, researchers being
interested in making forecasts to anticipate
the future evolution. It is important to know
the future values for variables like imports
and exports in order to ground the
government policy. Moreover, imports and
exports are components of GDP (gross
domestic product), the predictions for GDP
depending on the future values of exports
and imports.
The forecasting process should be
accompanied by the evaluation of these
1 Institute for Economic Forecasting, The Romanian Academy.
2 National University of Food Technologies, Kyiv, Ukraine.
predictions in terms of accuracy. For the
same variable, more alternative predictions
could be offered, but using the accuracy
criterion one should choose the best forecast
that is less affected by errors.
Therefore, this article is organized on
several sections that follow the logical
demarche. After the estimation of the
proposed models for imports and exports
indexes, some alternative forecasts are made.
The evaluation of these predictions is based
on several accuracy measures: root mean
square error, mean error, mean absolute
error, and U1 and U2 coefficients proposed
by Theil (1996).
2. Literature
Amano and Wirjanto (1994) considered
imports a linear function of domestic
demand and relative prices. The
coefficients are estimated with co-
integration techniques and generalized

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