Estimation of portfolio efficiency considering social responsibility: evidence from the multi-horizon diversification DEA

With the introduction of concept social responsibility investment/green investment, more and investors have realized importance such which has prompted portfolio managers to comprehensively consider both financial non-financial performance portfolios in different time horizons. DEA (Data Envelopment Analysis), as a data-driven evaluation approach, been widely used portfolios. However, existing studies are mostly limited single-horizon problems, indicators indicators, while ignoring impact ( e.g. , indicators). More importantly, input-output process multi-horizon framework also needs be clarified. In this paper, we first define from perspective, then propose corresponding stochastic output possibility sets based on returns indicators. We use expectation variance measures derive deterministic estimation above sets, where expectations variances all regarded outputs. construct diversification models with without constraints. Finally, select 20 component stocks China ESG100 index illustrate difference between models, further discuss efficiency its ranking. The empirical results show that compared proposed can provide an improvement strategy balance Further, find greater ranking, especially when pay attention performance.