Enhancement of value portfolio performance using data envelopment analysis

Purpose – The purpose of this paper is to examine the applicability of data envelopment analysis (DEA) as a basis of value portfolio selection criterion. Design/methodology/approach – The portfolios are composed of the comprehensive sample of Finnish non-financial stocks based on their DEA scale efficiency scores. The performance of portfolios is evaluated on the basis of average return and several risk-adjusted performance metrics. Moreover, the impact of holding period length on the results is examined by varying the portfolio reformation frequency from one to five years at annual frequency. Findings – The results show that the DEA scale efficiency scores add value to portfolio selection. Though outperformance of the DEA value portfolios in contrast to both comparable glamour portfolio and the stock market average is most evident for shorter (i.e. annual and biannual) holding periods, the absolute performance of the DEA value portfolio can be enhanced by using longer reformation intervals. Research limitations/implications – The sample of stocks is not large in spite of its comprehensiveness from the local stock market aspect. Future studies can apply DEA approach to other stock markets to examine whether the results are parallel to this study. Practical implications – The DEA is particularly useful as a multicriteria methodology in cases in which the number of stocks in the sample is large. Originality/value – This paper is the first attempt to form value portfolios using DEA models. The proposed methodology provides an interesting alternative to detect undervalued stocks by capturing several dimensions of relative value simultaneously. It provides also useful implications in portfolio management.