Page 297 - Contributed Paper Session (CPS) - Volume 6

CPS1935 Ahmed Oulad El Fakir

This table shows that the forecasts prepared by the HCP are of good quality
and close to reality. This is confirmed by the relative error which is very low
since its extreme values are only of the order of 0.3% below (for the years 2004
and 2006) or 1.4% above (only for the year 2016).
The graphical representation of these differences is given as follows:

Absolute and relative errors committed in forecasting
It should be noted that the greatest differences were recorded for the years
2004, 2006 and 2016. Thus, for the year 2006, the estimated growth rate was
5.2% but the circumstances have been different since this year saw a good
agricultural year and also had the positive effects of the Early Voluntary
Retirement announced in 2005. In contrast, the growth rate forecast in 2016
was overestimated and it was "called back to the order" from 2.6 % to 1.1%.
Does this difference in results reflect the limit of the forecast exercise?
Everyone knows that an economic forecast is obtained using a
mathematical model linking the set of macroeconomic variables in equations.
But, these ingredients (variables and equations linking these variables) differ
from one organism to another according to one's own vision to explain an
economic phenomenon to come and prepare the recipe to face it. Moreover,
human behavior is difficult to model and predict because it is subject to a
behavioral bias that is difficult to master and thus makes economic life very
uncertain. This explains the panoply of results obtained to explain the same
variable (the growth rate of GDP).

January of each year after the drafting of the Finance Law by the Ministry of Economy and
Finance and its adoption by the Two-Chamber Moroccan Parliament.
286 | I S I W S C 2 0 1 9