This README.txt file was generated on <2023-11-27> by <Ana Garcia-Bernabeu>

-------------------
GENERAL INFORMATION
-------------------

Title of Dataset: <Real-world datasets for sustainable portfolio selection>

Author Information <Para cada autor se debe poner el nombre, afiliación, dirección, email y ORCID>

	Principal Investigator:       Ana Garcia-Bernabeu, Department of Economy and Social Sciences - Universitat Polit�cnica de Val�ncia, Spain, angarber@upv.es, 0000-0003-3181-7745.
	Associate or Co-investigator: Adolfo Hilario-Caballero, Institute of Control Systems and Industrial Computing (ai2) - Universitat Polit�cnica de Val�ncia, Spain, ahilario@upv.es, 0000-0002-3237-8652.
	Associate or Co-investigator: Fabio Tardella, Department of Information Engineering  - University of Firenze, Italy, fabio.tardella@unifi.it, 0000-0002-0689-2981.
	Associate or Co-investigator: David Pla-Santamaria, Department of Economy and Social Sciences - Universitat Polit�cnica de Val�ncia, Spain, dplasan@upv.es, 0000-0003-1563-3997.



Date of data collection: <2021-01-18>

Geographic location of data collection: <Alcoy, Alicante, Comunidad Valenciana, Spain>.  < Latitud: N38¡42'19.62"- Longitud: O0¡28'27.55"> 


Information about funding sources or sponsorship that supported the collection of the data: <Agencia financiadora, nombre y código de proyecto vinculado a los datos>

General description: <We provide several real-world datasets for portfolio selection from three fund managers. The datasets contain monthly returns and ESG risk scores for each asset during nine years, from January 2011 to December 2019. We also provide, for these datasets, the Pareto fronts and Pareto sets corresponding to the Mean-Variance-ESG non-domintaed solutions computed by using a multiobjective genetic algorithm. >

Keywords: <Portfolio selection, Socially Responsible Investments, ESG, multiobjective genetic algorithm>


--------------------------
SHARING/ACCESS INFORMATION
-------------------------- 

Open Access to data: <Open = datos accesibles en abierto> 

Date end Embargo: <En caso de acceso abierto con embargo es necesario poner la fecha en la que los datos pasarán a estar en acceso abierto. Formato sugerido YYYY-MM-DD>

Licenses/restrictions placed on the data, or limitations of reuse: <Open Data Commons> (ODC-By, ODC-ODbL, etc.)... Más información en https://biblioguias.webs.upv.es/bg/index.php/es/datos-de-investigacion#DD>

Citation for and links to publications that cite or use the data: <Introducir enlace de las publicaciones, preferiblemente del DOI u otro identificador persistente>

Links/relationships to previous or related data sets: <Introducir el enlace de los datos, preferiblemente del DOI u otro identificador persistente>

Links to other publicly accessible locations of the data: <Introducir el DOI de los datos, preferiblemente del DOI u otro identificador persistente>



--------------------
DATA & FILE OVERVIEW
--------------------

File list: < File 1 - Dataset_SRI.xlxs, contains monthly returns and ESG risk scores for the set of assets provided by: Sheet- FM1 - Fund Manager 1, Sheet- FM2 - Fund Manager 2,Sheet- FM3 - Fund Manager 3. File 2 - Pareto_fronts.xlxs, contains the Pareto fronts including the Mean, Variance and ESG risk for FM1, FM2 and FM3 as well as the Pareto set that corresponds to these optimal solutions>

Relationship between files: < Pareto_fronts.xlxs is obtained when implementing a genetic algorithm using the in-sample time period form January 2011 to December 2016 >
Type of version of the dataset: <Dataset_SRI.xlsx are original values, while Pareto_fronts.xlxs are date generated for a given specification of the algorithm. >

Total size: <Dataset_SRI.xlxs- 93KB,  and Pareto_fronts.xlxs - 229KB>


--------------------------
METHODOLOGICAL INFORMATION
--------------------------

Description of methods used for collection/generation of data: <The Pareto fronts and Pareto sets have been generated using the ev-MOGA algorithm due to  M. Herrero, ÒEv-MOGA multiobjective evolutionary algorithm,Ó 2017, https://es.mathworks.com/matlabcentral/ fileexchange/31080-ev-moga-multiobjective-evolutionary-algorithm.

Methods for processing the data: <ev-MOGA, size of the main population 50000, auxiliary population 500, number of boxes 100 >

Software- or Instrument-specific information needed to interpret the data, including software and hardware version numbers: The proposed approach was coded in Matlab R2022b, and it was executed on an Intel i7 3.20 GHz, 20 GB (2667 MHz DDR4) of RAM.



Standards and calibration information, if appropriate:

Environmental/experimental conditions:

Describe any quality-assurance procedures performed on the data:



--------------------------
DATA-SPECIFIC INFORMATION <Crear secciones para cada archivo o conjunto de datos, según proceda>
--------------------------

Number of variables: 2-(returns and ESG risk scores) 

Number of cases/rows: FM1 - 35 assets, FM2 - 14 assets, FM3 - 16 assets

Variable list, defining any abbreviations, units of measure, codes or symbols used: monthly returns, 
   
Missing data codes:

Specialized formats or other abbreviations used: