Application of Kendall’s W Coefficient to Identify Groups of Statistically Related Variables
Marina Zhuravlyova, Vitalii Vedishchev · 2021 3rd International Conference on Control Systems, Mathematical Modeling, Automation and Energy Efficiency (SUMMA) · 2021
Original algorithm for identifying groups of statistically interrelated variables based on Kendall’s W statistics is considered. A formal description of the algorithm and assessing of the computational complexity are presented. The results of implementation of proposed algorithm for data of manufacturing process are shown. Comparison of the results obtained with the results of calculating Pearson and Spearman correlations for the same data is carried out.