Abstract
Within the context of Sustainable Development Goals, progress towards Target 12.3 can be measured and monitored with the Food Loss Index. A major challenge is the lack of data, which dictated many methodology decisions. Therefore, the objective of this work is to present a possible improvement to the modeling approach used by the Food and Agricultural Organization in estimating the annual percentage of food losses by country and commodity. Our proposal combines robust statistical techniques with the strict adherence to the rules of the official statistics. In particular, the case study focuses on cereal crops, which currently have the highest (yet incomplete) data coverage and allow for more ambitious modeling choices. Cereal data is available in 66 countries and 14 different cereal commodities from 1991 to 2014. We use the annual food loss as response variable, expressed as percentage over production, by country and cereal commodity. The estimation work is twofold: it aims at selecting the most important factors explaining losses worldwide, comparing two Bayesian model selection approaches, and then at predicting losses with a Beta regression model in a fully Bayesian framework.
References
1.
Barbieri, M.M., and J.O. Berger. 2004. “Optimal predictive model selection.” The annals of statistics , 32(3):870–897. DOI: https://doi.org/10.1214/009053604000000238.
2.
Bhadra, A., J. Datta, N.G. Polson, and B. Willard. 2019a. “Lasso meets horseshoe: A survey.” Statistical Science , 34(3):405–427. DOI: https://doi.org/10.1214/19-STS700.
3.
Bhadra, A, J. Datta, N.G. Polson, and B.T. Willard. 2019b. “The horseshoe-like regularization for feature subset selection.” Sankhya B. DOI: https://doi.org/10.1007/s13571-019-00217-7.
4.
Boxall, R.A. 1986. “A critical review of the methodology for assessing farm-level grain losses after harvest,” (working paper, Natural Resources Institute, University of Greenwich). Available at: http://gala.gre.ac.uk/10793/1/Doc-0235.pdf (accessed January 2020).
5.
Breiman, L. 2001. “Random forests.” Machine learning , 45(1):5–32. DOI: https://doi.org/10.1023%2fA:1010933404324.
6.
Carvalho, C.M., N.G. Polson, and J.G. Scott. 2009. “Handling sparsity via the horseshoe” Artificial Intelligence and Statistics : 73–80. Available at: http://proceedings.mlr.press/v5/carvalho09a.html (accessed January 2020).
7.
Carvalho, C.M., N.G. Polson, and J.G. Scott. 2010. “The horseshoe estimator for sparse signals.” Biometrika , 97(2):465–480. Available at: http://www.jstor.org/stable/25734098 (accessed January 2020).
8.
Celeux, G., F. Forbes, C.P. Robert, and D.M. Titterington. 2006. “Deviance information criteria for missing data models.” Bayesian analysis , 1(4):651–673. DOI: https://doi.org/10.1214/06-BA122.
9.
Cihan, P. 2018. “A comparison of five methods for missing value imputation in data sets.” International Scientific and Vocational Studies Journal , 2(2):80–85. Available at: https://dergipark.org.tr/en/download/article-file/614943 (accessed January 2020).
10.
Datta, J., and J.K. Ghosh. 2015. “In search of optimal objective priors for model selection and estimation.” Current Trends in Bayesian Methodology with Applications , 225. DOI: https://doi.org/10.1201/b18502-12.
11.
Fabi, C., A. English, M. Mingione, and G. Jona Lasinio. 2018. “Global foodloss index.” FAO, Rome, Sdg 12.3. 1. Available at: http://www.fao.org/3/CA2640EN/ca2640en.pdf (accessed May 2020).
12.
Ferrari, S., and F. Cribari-Neto. 2004. “Beta regression for modelling rates and proportions.” Journal of Applied Statistics , 31(7):799–815. DOI: https://doi.org/10.1080/0266476042000214501.
13.
FAO (Food and Agriculture Organization). 1980. Assessment and collection of data on post-harvest food-grain losses . Available at: http://www.fao.org/3/ca6157en/CA6157EN.pdf (Accessed January 2020).
14.
FAO (Food and Agriculture Organization). 2014. SAVE FOOD: Global Initiative on Food Loss and Waste Reduction. Definitional framework of food loss . Available at: http://www.fao.org/3/a-at144e.pdf, (accessed January 2020).
15.
FAO (Food and Agriculture Organization). 2019. Moving forward on food losses and waste reduction . Available at: http://www.fao.org/3/ca6030en/ca6030en.pdf (accessed January 2020).
16.
FAOSTAT (Food and Agricultural Organization Statistics Division). 2016. The United Nations Food and Agriculture Organization Database. Database, 2016. Available at: http://faostat3.fao.org/home/E (accessed February 2021).
17.
Franzin, A., F. Sambo, and B. Di Camillo. 2016. “bnstruct: an r package for bayesian network structure learning in the presence of missing data.” Bioinformatics , 33(8):1250–1252. DOI: https://doi.org/10.1093/bioinformatics/btw807.
18.
Gelman, A., J. Hwang, and A. Vehtari. 2014. “Understanding predictive information criteria for bayesian models.” Statistics and computing , 24(6):997–1016. DOI: https://doi.org/10.1007/s11222-013-9416-2.
19.
George, E.I., and R.E. McCulloch. 1993. “Variable selection via Gibbs sampling.” Journal of the American Statistical Association , 88(423):881 – 889. DOI: https://doi.org/10.1080/01621459.1993.10476353.
20.
Groll, A., J. Hambuckers, T. Kneib, and N. Umlauf. 2019. “Lasso-type penalization in the framework of generalized additive models for location, scale and shape.” Computational Statistics & Data Analysis , 140:59–73. DOI: https://doi.org/10.1016/j.csda.2019.06.005.
21.
Grünberger, K. (2013). Imputation of Loss Ratios for SUA/FBS Commodities . A Technical Report produced as part of the Food and Agriculture Organization of the United Nations CapEx Project. Available at: http://www.fao.org/3/i4315e/I4315E.pdf (accessed February 2021).
22.
Gustavsson, J., C. Cederberg, U. Sonesson, R. van Otterdijk, and A. Meybeck. 2011. Global food losses and food waste. Technical report, Food and Agricultural Organization (FAO). Available at: http://www.fao.org/3/a-i2697e/index.html (accessed July 2020).
23.
Harris, K.L., and C.J. Lindblad. 1978. Postharvest Grain Loss Assessment Methods . Available at: http://pdf.usaid.gov/pdf_docs/PNAAG842.pdf (accessed July 2020).
24.
International Energy Agency (IEA). 2019. Statistics resources – balance definitions. Available at: https://www.iea.org/statistics/resources/balancedefinitions/ (accessed November 2019).
25.
Ijarchelo, S.M., K. Afereydoon, and L. Zamanzadeh. 2016. “Bayesian variable selection under collinearity of parameters.” Reserach journal of applied sciences , 11:428–438. DOI: https://doi.org/10.3923/rjasci.2016.428.438.
26.
Kuiper, M., and H.D. Cui. 2020. “Using food loss reduction to reach food security and environmental objectives – a search for promising leverage points.” Food Policy : 101915. DOI: https://doi.org/10.1080/01621459.1988.10478694.
27.
Kuo, L., and B. Mallick. 1998. “Variable selection for regression models.” Sankhyā: The Indian Journal of Statistics, Series B , 65–81. Available at: https://www.jstor.org/stable/pdf/25053023.pdf?refreqid=excelsior%3A325e1cb3fa930037a8f90bdab30b9ac5 (accessed February 2021).
28.
Mitchell, T.J., and J.J. Beauchamp. 1988. “Bayesian variable selection in linear regression.” Journal of the American Statistical Association , 83(404):1023–1032. DOI: https://doi.org/10.1080/01621459.1988.10478694.
29.
O’Hara, R.B., and M. J Sillanpää. 2009. “A review of Bayesian variable selection methods: what, how and which.” Bayesian analysis , 4(1):85–117. DOI: https://doi.org/10.1214/09-BA403.
30.
Plummer. M. 2003. “JAGS: A program for analysis of Bayesian graphical models using Gibbs sampling.” In Proceedings of the 3rd International Workshop on Distributed Statistical Computing (DSC 2003). March 20–22, 2003, Vienna, Austria. 2003. DOI: https://www.r-project.org/conferences/DSC-2003/Proceedings/.
31.
Polson, N.G., and J.G. Scott. 2010. “Shrink lobally, act locally: Sparse bayesian regularization and prediction.” Bayesian statistics , 9(501-538):105. DOI: https://doi.org/10.1093/acprof:oso/9780199694587.003.0017.
32.
Polson, N.G., and J. G Scott. 2012. “Local shrinkage rules, lèvy processes and regularized regression.” Journal of the Royal Statistical Society: Series B (Statistical Methodology) , 74(2):287–311. DOI: https://doi.org/10.1111/j.1467-9868.2011.01015.x.
33.
Sinharay, S., H.S. Stern, and D. Russell. 2001. “The use of multiple imputation for the analysis of missing data.” Psychological methods , 6(4):317. DOI: https://doi.org/10.1037/1082-989X.6.4.317.
34.
Spiegelhalter, D.J., N.G. Best, B.P. Carlin, and A. van der Linde. 2014. “The deviance information criterion: 12 years on.” Journal of the Royal Statistical Society: Series B (Statistical Methodology) , 76(3):485–493. DOI: https://doi.org/10.1111/rssb.12062.
35.
Stekhoven, D.J., and P. Bühlmann. “Missforest–non-parametric missing value imputation for mixed-type data.” Bioinformatics , 28(1):112–118. DOI: http://dx.doi.org/10.1093/bioinformatics/btr597.
36.
Terenin, A.A., S. Dong, and D. Draper. 2019. “GPU-accelerated Gibbs sampling: a case study of the horseshoe probit model.” Statistics and Computing , 29(2):301–310. DOI: https://doi.org/10.1007/s11222-018-9809-3.
37.
Tibshirani, R. 1996. “Regression shrinkage and selection via the lasso.” Journal of the Royal Statistical Society: Series B (Methodological) , 58(1):267–288. Available at: www.jstor.org/stable/2346178 (accessed July 2020).
38.
Transforming our world: the 2030 Agenda for Sustainable Development. 2015. Avaliable at: https://sdgs.un.org/2030agenda (accessed February 2021).
39.
Vehtari, A., A. Gelman, and J. Gabry. 2017. “Practical Bayesian model evaluation using leaveone-out cross-validation and waic.” Statistics and computing , 27(5):1413–1432. DOI: https://doi.org/10.1007/s11222-016-9696-4.
40.
Waljee, A.K., A. Mukherjee, A.G. Singal, Y. Zhang, J. Warren, U. Balis, J. Marrero, J. Zhu, and P.DR. Higgins. 2013. “Comparison imputation methods for missing laboratory data in medicine.” BMJ open , 3(8). DOI: https://doi.org/10.1136/bmjopen-2013-002847.
41.
Watanabe, S. 2010. “Asymptotic equivalence of bayes cross validation and widely applicable information criterion in singular learning theory.” Journal of Machine Learning Research , 11:3571 – 3594. Avaliable at: http://jmlr.org/papers/v11/watanabe10a.html (accessed July 2020).
42.
White, I.R., P. Royston, and A.M. Wood. 2011. “Multiple imputation using chained equations: issues and guidance for practice.” Statistics in medicine , 30(4):377–399. DOI: https://doi.org/10.1002/sim.4067.
43.
Wu, L. 2009. Mixed Effects Models for Complex Data. CRC Press.