RT journal article
T1 A Study of the Colombian Stock Market with Multivariate Functional Data Analysis (FDA)
A1 Rodríguez Cuadro, D.
A1 Pérez Plaza, Sonia María
A1 Castaño Martínez, Antonia
A1 Fernández Palacín, Fernando
A2 Estadística e Investigación Operativa
K1 functional data analysis
K1 stock market
K1 volatility
K1 unctional principal component analysis
K1 k-means clustering
AB In this work, Functional Data Analysis (FDA) is used to detect behavioral patternsin the Bolsa de Valores de Colombia (BVC) in reaction to the global crises caused by COVID-19 and the war in Ukraine. The oil price fluctuation curve is considered a covariate. TheFDA’s distinctive ability is to represent stock values as smooth curves that evolve overtime and provide new insights into the dynamics of the BVC. The methodology makesuse of functional multivariate techniques applied to the smoothed curves of the closingprices of the main stocks of the BVC. The results show that the correlations of the oil curvewith the average market curve change from almost null or low in the global period toextremely significant in time windows immediately after the beginnings of COVID-19 andthe war in Ukraine, respectively. On the other hand, the velocity curves, which are usedto evaluate the stock market volatility, show a pattern of synchronization of companies inthe crisis periods. Furthermore, in these crisis periods, the companies in BVC showed ahigh synchronization with the Brent crude oil price. In conclusion, this work shows theusefulness of the FDA as a complement to time series analysis in the study of stock markets.The results of this research could be of interest to academic researchers, financial analysts,or institutions.
PB MDPI
SN 2227-7390
YR 2025
FD 2025-03-05
LK http://hdl.handle.net/10498/36935
UL http://hdl.handle.net/10498/36935
LA eng
DS Repositorio Institucional de la Universidad de Cádiz
RD 10-nov-2026