Crop Yield Insurance Analysis for Turkey: Spatiotemporal Dependence


Şimşek G., Yildirak Ş. K.

in: Quantitative Risk Management in Agricultural Business, Hirbod Assa,Peng Liu,Simon Wang, Editor, Springer Nature Switzerland Ag, Zürich, pp.173-196, 2025

  • Publication Type: Book Chapter / Chapter Research Book
  • Publication Date: 2025
  • Publisher: Springer Nature Switzerland Ag
  • City: Zürich
  • Page Numbers: pp.173-196
  • Editors: Hirbod Assa,Peng Liu,Simon Wang, Editor
  • Hacettepe University Affiliated: Yes

Abstract

Farming is among the most vulnerable segments of society due to the source of the income that is highly dependent on environmental risks. To maintain their production, farmers, who are critical components of agricultural production, need to protect themselves against production risks. For farmers to continue agriculture, it is crucial to provide insurance policies that at the very least protect their current income. Therefore, crop yield insurance has been discussed in this study. When a crop yield falls short of a predetermined threshold, crop yield insurance compensates for the resulting yield loss. This insurance product holds a prominent position among other agricultural insurances because yield insurance, which aims to keep agricultural production at a specific level, maintains sustainability in the ecosystem. Through the spatiotemporal modeling of crop yields and yield insurance, the impact of climate change, a major problem for agricultural insurance, has also been addressed. For the conditional crop yield distribution in this study, a hierarchical Bayesian technique is employed to characterize the spatiotemporal dependence. Wheat yield statistics from the years 2004 to 2022 were used for a total of 47 districts that are part of Ankara and Konya, which are at the top of the list in terms of wheat production volume. Premium rates have been obtained for the region, province, and chosen districts using the preferred model in accordance with model selection and performance criteria, and the results are presented. The R-INLA package program is used to perform all statistical analyses for this study.