Cardinality-constrained approach: Small portfolios breakthrough in the Philippine market from January 2015 to December 2019

• Main Authors: Cabador, Judely Ann Calipusan, Calo-oy, Clarissa Lingat, Duldulao, Krisma Allu Gapasin, Zamora, Juliene Faye Palmares
• Format: text
• Language: English
• Published: Animo Repository 2021
• Subjects: Portfolio management—Philippines; Stocks—Philippines; Corporate Finance; Finance and Financial Management
• Online Access: https://animorepository.dlsu.edu.ph/etdb_finman/16; https://animorepository.dlsu.edu.ph/cgi/viewcontent.cgi?article=1000&context=etdb_finman
• Institution: De La Salle University

Diversification of assets usually implies that a portfolio should contain a large number of assets. However, this paper constructed small but optimal portfolios called cardinality-constrained portfolios that can attain returns as good as large portfolios. This paper specifically constructed 3, 5, and 10-stock portfolios using the 26 consistent stocks of the benchmark Philippine Stock Exchange index (PSEi) for periods 2015-2019. Three statistical tools were applied to build these portfolios and evaluate their performance: the graduated non-convexity method, shrinkage estimator and stationary bootstrap. Aside from rebalancing the cardinality-constrained portfolios weekly and monthly, a cross-sectional analysis for their three- and five- year performances was conducted. Results showed that cardinality-constrained portfolios were able to beat the market based on mean return and Sharpe ratio. Specifically, for the 5-year results of mean returns, the best performing was the 3-stock for weekly rebalancing (0.209) and the 5-stock for monthly rebalancing (0.557). For the 3-year results of mean returns, the best performing was the 10-stock for weekly rebalancing (0.147) and the 3-stock for monthly rebalancing (-0.065). Meanwhile, for the 5-year results of the Sharpe ratio, the best performing was the 5-stock for weekly rebalancing (0.108) and the 3-stock for monthly rebalancing (0.042). For the 3-year results of the Sharpe ratio, the best performing was the 10-stock for weekly rebalancing (0.073) and the 5-stock for monthly rebalancing (-0.022).