Naslov
Uncertainty of forecasted new service market capacity obtained by logistic model

Autori
Sokele, Mladen

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
ISF 2008 Proceedings / Garcia-Ferrer, Antonio ; Hamoudia, Mohsen - Nica : International Institute of Forecasters, USA, 2008, 17.1-17.15

Skup
28th Annual International Symposium on Forecasting (ISF) Information Communication Technology (ICT) Forecasting in a Digital World

Mjesto i datum
Nica, Francuska, 22.05.2008. - 25.05.2008

Vrsta sudjelovanja
Pozvano predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Market capacity forecasting; Logistic model; Uncertainty intervals

Sažetak
Logistic model of growth is widely used model for technological and market development forecasting because of its many useful properties. In telecommunications, logistic model is used as a quantitative forecasting method for the new service market adoption when interaction with other services can be neglected. Growth forecasting relies on hypothesis that extrapolation of model, which is fitted to known data points, will be valid in the perceivable future. Thus, parameters of the model as well as forecasted values are sensitive to the accuracy of input data points. In general, logistic model parameter determination requires the application of an iterative numerical method, which complicates direct assessment of model sensitivity to uncertainty of input data. Namely, the uncertainty in the determination of telecommunications service market capacity is of great concern to operators. Presented analytical procedure for direct logistic model parameter determination in case of equidistant time points gives valuable basis for the full sensitivity analysis. Expressions and contour graphs showing dependence of forecasted market capacity on uncertainty of input data are obtained by the total differentiation approach. In addition, required conditions and analysis of practical cases with input data uncertainty influence on forecasting results are presented.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Informacijske i komunikacijske znanosti

POVEZANOST RADA