Prediction of Human Development from Environmental Indicators
The Sustainable Development Goals (SDG) list the objectives and targets that should be addressed to solve the global issues regarding sustainable development. They encompass the social, economic and environmental dimensions and search for solutions that are able not only to monitor but also to control the operational indicators attributed to each objective. It is expected that many of these indicators are associated to each other and the accurate understanding of these correlations allows to build predictive statistical models that can improve the monitoring and controlling of variables. It would increase the rate of success in achieving the SDG. This study tested a linear multivariate model able to predict the human development index based on environmental variables which are related to SDG 3 (health), 4 (education), 8 (sustainable economic growth and decent work) and 15 (protect, restore and promote sustainable use of terrestrial ecosystems). We fitted the models using the Linear Discriminant Analysis and Best Subset Selection applied to a Linear Multivariate Regression. The model predictive ability was assessed by R2 and cross-validation (CV). The results showed that exposure to unsafe sanitation, access to drinking water, tree cover loss, unsafe water quality, wastewater treatment level, and household air pollution are excellent predictors of human development index of a population, with R2 = 0.94 and 10-fold CV Mean Squared Error equal to 0.0014. This tool can help stakeholders to monitor and control indicators attributed to good health and well-being, quality education, clean water and sanitation, decent work and economic growth, sustainable cities and communities and life on land sustainable development goals.
KeywordsEnvironmental Performance Index Human development index Linear Discriminant Analysis Machine learning Cross-validation
- Hsu, A., Emerson, J., Levy, M., de Sherbinin, A., Johnson, L., Malik, O., et al. (2016). Environmental performance index. New Haven, CT: Yale University. www.epi.yale.edu.
- Hughes, B., Irfan, M., Moyer, J., Rothman, D., & Solórzano, J. (2011). Forecasting the impacts of environmental constraints on human development. Human Development Research Paper 8. New York: UNDP–HDRO.Google Scholar
- James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning: With applications in R (Corr. 4th). New York: Springer.Google Scholar
- Kovacevic, M. (2011). Review of HDI critiques and potential improvements. UNDP, Human Development Reports 2010/33.Google Scholar
- Lumley, T. (2013). Package ‘leaps’. http://cran.r-project.org/web/packages/leaps/leaps.pdf
- Pinter, L., Hardi, P., & Bartelmus, P. (2005). Sustainable development indicators: Proposals for a way forward (pp. 1–11). Discussion paper prepared under a consulting agreement on behalf of the UN Division for Sustainable Development, IlSD, Winnipeg, CDN.Google Scholar
- R Core Team (2016). R: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. https://www.R-project.org/.
- UN. (2015). Transforming our world: The 2030 agenda for sustainable development. Resolution adopted by the General Assembly on September 25, 2015, A/RES/70/1. Geneva: United Nations.Google Scholar
- UNDP. (2015). Human development report [Internet]. New York: PBM Graphics. http://hdr.undp.org/sites/default/files/2015_human_development_report.pdf.
- UN Economic and Social Council. (2016). Report of the inter-agency and expert group on sustainable development goal indicators. https://unstats.un.org/unsd/statcom/47th-session/documents/2016-2-IAEG-SDGs-E.pdf.
- UNSD. (2016). Third meeting of the Inter-Agency and Expert Group on the Sustainable Development Goal Indicators. Report, New York (UN Doc. ESA/STAT/AC.318/L.3). http://unstats.un.org/sdgs/files/meetings/iaeg-sdgs-meeting-03/3rd-IAEG-SDGs-Meeting-Report.pdf.