Analysing farmland rental rates using Bayesian geoadditive quantile regression

Abstract

Empirical studies on farmland rental rates so far have predominantly concentrated on modelling conditional means using spatial autoregressive models. While these models only focus on the central tendency of the response variable, quantile regression provides more detailed insight by modelling different points of the conditional distribution as a function of covariates. Based on data from the German agricultural census, this article contributes to the agricultural economics literature by modelling conditional quantiles of farmland rental rates semi-parametrically using Bayesian geoadditive quantile regression. Our results stress the importance of using semi-parametric regression models, as several covariates influence rental rates in an explicit non-linear way. Moreover, our analysis allows us to uncover potential heterogeneities of the estimated effects across the conditional distribution of rental rates. By explicitly modelling and visually presenting the spatial effects, we also provide additional insight into the spatial structure of German farmland rental rates.