Portfolio Selection Using Data Envelopment Analysis

• Main Authors: Bhagia, Vishal, Chiu, Colleen Monica K., Castillo, Paulynne, Raymundo, Roberto, Tanchuco, Joel Q.
• Format: text
• Published: Animo Repository 2021
• Subjects: Data Envelopment Analysis; Portfolio Optimization; DEA; Mean-Variance Portfolio; mean variance portfolio; MVP; Cross-Efficiency; Work, Economy and Organizations
• Online Access: https://animorepository.dlsu.edu.ph/res_aki/97; https://animorepository.dlsu.edu.ph/context/res_aki/article/1098/viewcontent/Portfolio_Selection_Using_Data_Envelopment_Analysis.pdf
• Institution: De La Salle University

There has been a growing interest in applying data envelopment analysis (DEA) as a non-parametric approach in portfolio optimization due to its flexibility in overcoming the limitations of the conventional mean-variance portfolio (MVP) model. Therefore, this study highlights the use of DEA as a portfolio selection tool that may encourage individuals to invest in the Philippine stock market for its ability to integrate any technical and fundamental factors. This study shows that the DEA model outperforms the MVP model in terms of risk-adjusted returns. However, the investor may need to change the model used to generate the highest returns because the investor may either hold a short-term or long-term investment. This study recommends that the investor does the following: (a) formulate short-term portfolios using the DEA model as it outperforms the MVP in the short-run and can provide for a versatile set of inputs and outputs that determine the optimal portfolio, and (b) formulate long-term portfolios using the MVP model as returns are mean-reverting in the long-run.