Page 238 - Contributed Paper Session (CPS) - Volume 6

CPS1907 Klaudia M. T. et al.

Figure 4 The actual and estimated value added of industry 61-63
1,500
1,000

500

40,000 0

20,000 -500
-1,000
0

-20,000

-40,000
05 06 07 08 09 10 11 12 13 14 15 16 17 18
Forecast for second difference of 61_63
Second difference of 61-63 Error of forecast
We argue based on this result, the gross value added of information and
telecommunication (q61_63) presents a good forecast efficiency. The results
show that the crisis in 2008-2009 and the recovery at the beginning of 2014
means a relative significant challenge for the forecast algorithm.

5. Discussion and Conclusion
The current applied flash estimation based on bottom-up concept with
ARIMA models faces significant problems in the case of rapid economic and
technological challenges and considerable multicollinearity of time series.
Both problem appear in the information and communication industry in
Hungary. The rapid changes in telecommunication technologies are followed
in the development of gross value added as well, therefore it is a great
motivation to improve a new model for estimation.
The applied TVC model is a kind of state space models and can handle the
nonlinearity in variables, although it uses linear equations. The fitted TVC
model utilizes the physical indicators for estimation process and follows the
structural changes of telecommunication. The predictive ability is appropriate,
it allows more accurate estimation for the industry. The disadvantage of the
method is that it cannot handle the changes in trend, although changes in
trend causes problems in ARIMA models as well.

227 | I S I W S C 2 0 1 9