Sonia Maria Perez-Plaza

Dr. Sonia Maria Perez-Plaza

orcid

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 May 2018 in Mathematical Geosciences
Verified
Journal Article
Closed access
PDF file
SPSonia Maria Perez-Plaza
Sonia Maria Perez-Plaza
FFFernando Fernández-Palacín
Fernando Fernández-Palacín
MDManuel Berrocoso Domínguez
Manuel Berrocoso Domínguez
RPRaúl Páez
Raúl Páez
BRBelén Rosado
Belén Rosado

5 March 2025 in Mathematics
Verified
Journal Article
Open access
PDF file
DRDeivis Rodríguez
Deivis Rodríguez
SPSonia Maria Perez-Plaza
Sonia Maria Perez-Plaza
ACAntonia Castaño-Martínez
Antonia Castaño-Martínez
FPFernando Fernández Palacín
Fernando Fernández Palacín
In this work, Functional Data Analysis (FDA) is used to detect behavioral patterns in 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. The FDA’s distinctive ability is to represent stock values as smooth curves that evolve over time and provide new insights into the dynamics of the BVC. The methodology makes use of functional multivariate techniques applied to the smoothed curves of the closing prices of the main stocks of the BVC. The results show that the correlations of the oil curve with the average market curve change from almost null or low in the global period to extremely significant in time windows immediately after the beginnings of COVID-19 and the war in Ukraine, respectively. On the other hand, the velocity curves, which are used to evaluate the stock market volatility, show a pattern of synchronization of companies in the crisis periods. Furthermore, in these crisis periods, the companies in BVC showed a high synchronization with the Brent crude oil price. In conclusion, this work shows the usefulness 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.

21 September 2021 in Earth, Planets and Space
Verified
Journal Article
Open access
PDF file
SPSonia Maria Perez-Plaza
Sonia Maria Perez-Plaza
MBManuel Berrocoso
Manuel Berrocoso
BRBelén Rosado
Belén Rosado
GPGonçalo Prates
Gonçalo Prates
FFFernando Fernández-Palacín
Fernando Fernández-Palacín
On 10 October 2011, a submarine eruption occurred in El Hierro island. Thus, the eruptive process in the Canary islands was reactivated after 40 years of inactivity. The main objective of this work is to evaluate, using Functional Data Analysis, how the surface deformation phenomenon explains the seismic–volcanic activity in the island. The GNSS-GPS data are from the FRON (GRAFCAN) station, located in Frontera. These data measure, each 4 h, the distance between the FRON station and the reference station LPAL (La Palma island) from August, 2010 to December, 2013. In this study a functional correlation measure is employed to establish the relation between the deformation curve and the curve of cumulative energy released. The period of time analysed has been divided into four phases to avoid the mix of phenomena. For each phase, the correlation measure and the time lag between deformation curve and the curve of cumulative energy released have been estimated. These values show a strong relation between these curves. With respect to time lag period, the only significant lag, of about 1 month, occurred in Phase 1, which was after a long period without seismic activity. The later phases had very short, insignificant, lags. After a long period without seismic and volcanic activity in El Hierro island, the time lag between the deformation process and the beginning of the seismic activity takes approximately 1 month. In a similar situation a method to predict in real time the beginning of the seismic activity is proposed. This method, based on the changes produced in the derivative curves when there is a rapid descent in the deformation curve, could activate a warning system approximately 13 days before the beginning of seismicity.

7 February 2014 in International Journal of Engineering, Science and Technology
Verified
Journal Article
Open access
PDF file
AAAj Arriaza-Gómez
Aj Arriaza-Gómez
FFF Fernández-Palacin
F Fernández-Palacin
MMM Muñoz-Marquez
M Muñoz-Marquez
SPSonia Maria Perez-Plaza
Sonia Maria Perez-Plaza
In the paper “A hybridized K-means clustering approach for high dimensional dataset” Dash, Mishra, Rash and Acharya have presented a new version of the k-means algorithm. In it, principal components analysis (PCA) was used before applying the kmeans algorithm with a new initialization method. The authors compare the results obtained by using the HKMCA and PCA with the results of the original k-means, but a direct comparison is not valid as this paper shows. Keywords: Cluster analysis, K-means Algorithm, Principal Component Analysis, Hybridized K-means algorithmInternational Journal of Engineering, Science and Technology Vol. 6, No. 1, 2014, pp. 20-26