نوع مقاله : مقاله پژوهشی
عنوان مقاله English
نویسندگان English
Abstract
Partial correlation is a metric used to measure the dependence between two random variables while controlling for the influence of other variables. In this study, interaction networks based on Pearson correlation and partial correlation are compared in the context of the Tehran Stock Exchange. The results show that the minimum spanning tree (MST) constructed from Pearson correlation exhibits a star-like structure with a dominant central node, whereas the MST derived from partial correlation has a more uniform structure. This contrast indicates that ignoring the effect of common variables may lead to misleading interpretations of network structure. Furthermore, the analysis of the generalized network dimension as a function of the parameter q reveals that networks based on partial correlation are more effective in distinguishing between real and random data. Finally, the relationship between network dimension and Rényi entropy is examined in both types of networks, and their distinctive characteristics are analyzed.
Interaction network, Minimum Spanning Tree, Pearson correlation, Partial correlation, Tehran Stock Exchange
Keywords
Interaction network, Minimum Spanning Tree, Pearson correlation, Partial correlation, Tehran Stock Exchange
کلیدواژهها English