Результаты поиска по 'predictability':
Найдено статей: 92
  1. Timiryanova V.M., Lakman I.A., Larkin M.M.
    Retail forecasting on high-frequency depersonalized data
    Computer Research and Modeling, 2023, v. 15, no. 6, pp. 1713-1734

    Technological development determines the emergence of highly detailed data in time and space, which expands the possibilities of analysis, allowing us to consider consumer decisions and the competitive behavior of enterprises in all their diversity, taking into account the context of the territory and the characteristics of time periods. Despite the promise of such studies, they are currently limited in the scientific literature. This is due to the range of problems, the solution of which is considered in this paper. The article draws attention to the complexity of the analysis of depersonalized high-frequency data and the possibility of modeling consumption changes in time and space based on them. The features of the new type of data are considered on the example of real depersonalized data received from the fiscal data operator “First OFD” (JSC “Energy Systems and Communications”). It is shown that along with the spectrum of problems inherent in high-frequency data, there are disadvantages associated with the process of generating data on the side of the sellers, which requires a wider use of data mining tools. A series of statistical tests were carried out on the data under consideration, including a Unit-Root Test, test for unobserved individual effects, test for serial correlation and for cross-sectional dependence in panels, etc. The presence of spatial autocorrelation of the data was tested using modified tests of Lagrange multipliers. The tests carried out showed the presence of a consistent correlation and spatial dependence of the data, which determine the expediency of applying the methods of panel and spatial analysis in relation to high-frequency data accumulated by fiscal operators. The constructed models made it possible to substantiate the spatial relationship of sales growth and its dependence on the day of the week. The limitation for increasing the predictive ability of the constructed models and their subsequent complication, due to the inclusion of explanatory factors, was the lack of open access statistics grouped in the required detail in time and space, which determines the relevance of the formation of high-frequency geographically structured data bases.

    Keywords: cash registers data, retail scanner data, depersonalized high-frequency data, fiscal data operator, spatial regression on panel data.
  2. Korenkov V.V., Nechaevskiy A.V., Ososkov G.A., Pryahina D.I., Trofimov V.V., Uzhinskiy A.V.
    Synthesis of the simulation and monitoring processes for the development of big data storage and processing facilities in physical experiments
    Computer Research and Modeling, 2015, v. 7, no. 3, pp. 691-698

    The paper presents a new grid and cloud services simulation system. This system is developed in LIT JINR, Dubna, and it is aimed at improving the efficiency of the grid-cloud systems development by using work quality indicators of some real system to design and predict its evolution. For these purpose, simulation program is combined with real monitoring system of the grid-cloud service through a special database. The paper provides an example of the program usage to simulate a sufficiently general cloud structure, which can be used for more common purposes.

    Keywords: grid computing, cloud computing, data storage, monitoring, optimization, simulation.
    Views (last year): 4. Citations: 6 (RSCI).
Pages: « first previous

Indexed in Scopus

Full-text version of the journal is also available on the web site of the scientific electronic library eLIBRARY.RU

The journal is included in the Russian Science Citation Index

The journal is included in the RSCI

International Interdisciplinary Conference "Mathematics. Computing. Education"

Website Copyright © 2009–2025 Institute of Computer Science