文章基本信息

标题：Data Reduction in Socio-Economic Studies
本地全文：下载
作者：V. V. Lypchuk ; О. M. Krupa
期刊名称：Статистика України
印刷版ISSN：2519-1853
电子版ISSN：2519-1861
出版年度：2017
期号：1(76)
页码：15-20
DOI：10.31767/su.1(76).2017.01.02
出版社：State Statistics Service of Ukraine, the National Academy of Statistics, Accounting and Audit (NASAA), the National Academy for Public Administration (NAPA) under the President of Ukraine
摘要：Показано, що ускладнення процесу прийняття рішень в умовах невизначеності та труднощі прогнозування динамічного розвитку різноманітних суспільних явищ спричинюють потребу у щоразу більшій кількості первинних даних, які нагромаджуються у великих обсягах. Обгрунтовано необхідність редукції даних як важливого етапу забезпечення достовірності та економічності проведення соціально- економічних досліджень. Розглянуто суть процесу редукції, подано етапи її здійснення та перелічено найбільш типові використовувані методи.↓Показано, что усложнение процесса принятия решений в условиях неопределенности и трудности прогнозирования динамического развития различных общественных явлений обусловливают потребность в растущем количестве первичных данных, которые накапливаются в больших объемах. Обоснована необходимость редукции данных как важного этапа обеспечения достоверности и экономичности проведения социально-экономических исследований. Рассмотрены суть процесса редукции, представлены этапы ее осуществления и перечислены наиболее типичные используемые методы.
其他摘要：The article is devoted to the problem of data reduction as an important step on the way of providing reliability and efficiency of socio-economic studies. Through the reduction the large amounts of raw data, generated from different sources, become more useful, convenient and clear for use. Meanwhile, the data reduction is not treated as a separate phase of studies in the national statistic practice. The aim of the article is to substantiate the importance of data reduction in economic studies and attempt to systematize and generalize the essence and components of the phase of data reduction as well as ways of their implementation. The study is based on methods of theoretical generalization, abstract and logic, analogy and others. The essence of data reduction is defined as the process of converting raw data into the pure form and reducing the number of units’ attributes (features), which are not significant to further analysis. In fact, this is part of the analysis involving selection of the data that are most important from the viewpoint of the study’s goals. The significance of data reduction in economic studies is outlined. It is found that it assures the validity of their results, reduces their time and costs, simplifies the representational complexity of the problem being addressed, eliminates the errors and redundant data from the investigated set, looses the requirements to calculation tools. The data resulting from reduction are much more informative. Many dependencies and relationships become more readable (visual). It is emphasized that reduction applies to the current data (on-line), as well as to historical data (off-line), contained in the already created databases. The phases of data reduction are described. They are: control of data collection, data editing, classification, data construction and grouping, coding and transmission (data transmission to the processing tools - computers). Data reduction techniques and methods most common in the global practice are shown. Future studies of data reduction problems are expected to focus on potential ways to implement its advanced methods in the domestic practice of statistical science. It will allow for enhancing significantly the speed and efficiency of economic analysis and the reliability of its results.
关键词：socio-economic studies; data reduction; substantial signs; techniques; stages; credibility; reduction methods.
其他关键词：socio-economic studies; data reduction; substantial signs; techniques; stages; credibility; reduction methods