ANALYZING THE RESULTS OF VERIFICATION OF SHORT-TERM FORECASTS OF FIGURES OF ACADEMIC RESEARCH AND INNOVATION SPHERE IN RUSSIA

On the basis of authors' programs and methodological approaches the automated system of calculating short-term forecast of the parameters in the field of research and innovation was developed. The article for the first time raised and solved the task of analyzing dynamic assessment of accuracy...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Igor B. Kolmakov, Olga V. Kitova, Aleksey V. Koltsov, Matvey V. Domozhakov
Formato: article
Lenguaje:RU
Publicado: Plekhanov Russian University of Economics 2017
Materias:
Acceso en línea:https://doaj.org/article/071b7cfcc6f0493f98e90fe6e5f29b42
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Descripción
Sumario:On the basis of authors' programs and methodological approaches the automated system of calculating short-term forecast of the parameters in the field of research and innovation was developed. The article for the first time raised and solved the task of analyzing dynamic assessment of accuracy and quality of retro-forecast parameters on the data for 2012, 2013 and 2014. Figures of the research and developments observed by the Rosstat were analyzed, as well as methodology of their shaping and trajectories of the data for 2004-2014. The authors found the reasons for restrictions and maximum possibilities of using regressive models of the forecast. A specialized complex of automated calculation accuracy and quality of retro-forecast. The research proved conclusions about qualitative restrictions of possibilities of econometric models and allowed to get qualitative assessment for parameters in the field of research and innovation. More than 80% of figures are forecasted successfully. As for other figures it is recommended to use alternative methods of forecast, for instance neuro-network. The use of verification in the automated system of calculating the comparative assessments could increase the speed, accuracy and quality of regressive equation adjustment.