Scientific Presentation
HOMECOMINGAbstract
Farm Accountancy Data Network (FADN) comes without the exact location. It is generally aggregated at NUTS-2 and without agro-climatic information. However, the assessment of weather-related farm production effects requires agro-climatic and spatial farm information. When analyzing effects of weather on agricultural outcomes with incorrect weather variables, we potentially introduce measurement errors in weather regressors (Li and Ortiz-Bobea, 2022). Existing downscaling approaches not available at temporal and spatial requirements for 15-year time-series and EU-wide analysis (Kempen et al., 2011). Therefore, we develop an open-access framework to downscale FADN farm-level data from NUTS-2 to NUTS-3 utilizing a Bayesian Highest Posterior Density Concept.
Meet alumni working on the development economics, practitioners and academics. I will present our work on the probabilistic spatial downscaling of FADN data.
