CLUSTER ANALYSIS APPROACHES TO ASSESSING THE FINANCIAL AND ECONOMIC ACTIVITIES OF ENTERPRISES
Journal Title: Комп’ютерне моделювання: аналіз, управління, оптимізація - Year 2019, Vol 1, Issue 1
Abstract
The article is devoted to the financial and economic assessment of the state of road transport enterprises in Ukraine through the use of multidimensional statistical cluster analysis. The article solved the task of analyzing the financial and economic state of enterprises in Ukraine for 2013-2017 years in the STATISTICA software product, taking into account financial and economic indicators. The study used four groups of indicators: liquidity, financial stability, business activity, profitability according to the financial statements of ten road transport enterprises. The use of the k-means method and the agglomerative tree-clustering method made it possible to divide enterprises into three clusters that reflect their state: crisis, pre-crisis, and non-crisis. When the k-means method is used, the initial centers of the clusters are specified by the method of sorting the distance and selecting observations at constant intervals, observations that maximize the initial distances between the clusters are chosen. The value of intergroup and average group variances, the F parameter and the level of significance were obtained, which made it possible to identify financial and economic indicators that are most informative for the clustering process using the k-means method. For clustering using the agglomerative tree-clustering method, the single link method was taken as the clustering rule. As a result of clustering, a dendrogram was constructed, which allows determining the clusters depending on the distance between them. This made it possible to analyse the characteristics of various aspects of the financial condition of the enterprise, to identify the parameters that form the initial clusters of the state of enterprises, and the clusters, which form separate clusters and gradually join the first clusters. A horizontal tree diagram of road transport enterprises has been built, which reflects enterprises that are closely shrinking with each other and form new clusters by combining. The clusters obtained using the method of k-means and agglomerative method of tree clustering were analyzed and conclusions regarding their state were drawn.
Authors and Affiliations
В. Р. Писарькова, Н. Ю. Науменко
Влияние дисперсности материалов на амплитуду акустических сигналов рабочей зоны
Работа посвящена исследованию влияния размера частиц газодисперсных потоков на амплитуду акустических сигналов, как в процессе измельчения, так и в процессе транспортировки сыпучих материалов. Экспериментально подтвержде...
Neural networks as a means of increasing the accuracy and efficiency of solving the optimization problems
A method is proposed for increasing the accuracy and efficiency of calculating constraint functions in problems of optimal design of structures exposed to aggressive media, with constraints on a given durability (operati...
Прогнозування чисельності окремих популяцій в одній екологічній зоні
В роботі розглядається моделювання процесів зміни динаміки популяцій під впливом різних факторів навколишнього середовища. Зроблено спробу спрогнозувати чисельність окремих популяцій в умовах нерівномірного розподілу вид...
Разработка критериев выбора рациональной калибровки бандажей валковых прессов
Показана актуальность и пути формирования научно-обоснованного метода определения рациональной калибровки валков брикетных прессов на основе анализа связей между параметрами калибровки бандажей и характеристиками процесс...
Modeling a rheological anomaly in aqueous aminosilicate solutions
A simple mathematical model of interrelated acid-base and polycondesation equilibria in aqueous solutions of aminosilicates (polysilicates of aliphatic amines, amino alcohols and other amino compounds) is proposed. The o...