MSc thesis subject: Spatially explicit crop variety portfolio management for smallholders

Uncertainty about weather conditions in the next growing season that will affect disease pressure and yield potential urges farmers to spread risks by selecting a mix of crop varieties. The composition of optimal crop variety mixes depends on spatially explicit abiotic and biotic conditions while it is also influenced by taste preferences, market demands, etc. that may vary in space.

The main objective of this thesis topic is to explore and assess the applicability of so-called variety portfolio management methods for smallholder farmers. Data are from large-N on-farm trials involving large groups of farmers who test and disseminate new varieties on their farms: Bioversity International’s “triadic comparisons of technologies” (van Etten et al., 2017).
Decision theoretic methods, such as regret minimization (Dembo and King, 1992) are commonly used for managing investment portfolios but application in the domain of crop varieties management is relatively novel (Sukcharoen and Leatham, 2016). Recent research (yet to be published) has shown the potential of the approach, though.


  • Explore and assess the applicability of variety portfolio management by regret minimization.
  • Derive required variety performance indicators from triadic comparisons of technologies.
  • Map local (a)biotic and socio-economic conditions.
  • Facilitate interpretation and decision making with appropriate reporting and visualization methods.


  • Dembo, R.S. and King, A.J., 1992. Tracking models and the optimal regret distribution in asset allocation. Applied Stochastic Models and Data Analysis, 8: 151-157
  • van Etten, J., E. Beza, L. Calderer, K van Duijvendijk, C. Fadda, B. Fantahun, Y.G. Kidane, J. van de Gevel, A. Gupta, D.K. Mengistu, D. Kiambi, P. Mathur, L. Mercado, S. Mittra, M. Mollel, J.C. Rosas, J. Steinke, J.G. Suchini, K. Zimmerer., 2017. First experiences with a novel farmer citizen science approach: Crowdsourcing participatory variety selection through on-farm triadic comparisons of technologies (tricot). Experimental Agriculture, Online.
  • Sukcharoen, K. and Leatham, D., 2016. Mean-variance versus mean-expected shortfall models: an application to wheat variety selection. Journal of Agricultural and Applied Economics, 48, 2: 148-172.


  • Strong analytical skills
  • Scripting skills

Theme(s): Modelling & visualisation, Empowering & engaging communities