Minimal Spanning Tree application to determine market correlation structure

Abstract

Determining the structure of market correlation is an important topic in theory and experiments. Under the impact of the Covid-19 pandemic, the market structure may be deformed. Therefore, this study examines the pandemic’s impact on the market structure. This study considered the correlation structure of the VN30 portfolio (including 30 stocks with the largest market capitalization); the collecting period is from July 28, 2000, to July 30, 2021. The data was divided into 02 phases before and after the pandemic. The Kruskal algorithm is implemented to determine the Minimal Spanning Tree (MST) structure to define the structure of market correlation. This study compared the change in the structure before and after the Covid-19 pandemic by structures’ mean of distances comparison. T-test results show that there are structural differences before and after the pandemic. Based on the research result, investors should change their risk management strategy to suit the market context because the previous structure has been changed.