LAUSR

Aim of study: To obtain spatial land valuing models using Geographic Information Systems (GIS), which collect spatial autocorrelation and improve the conventional models estimated by OLS (Ordinary Least Squares) to determine and quantify the factors explaining these values.Area of study: The Spanish Autonomous Community of Aragón, Spain.Material and methods: The mean land values per municipality and the land uses published by the Aragonese Statistics Institute were used, as well as the geographic, agricultural, demographic, economic and orographic characteristics of these municipalities. The Spatial Lag Model and the Spatial Error Model were compared with OLS in general terms and for uses.Main results: The statistics (R2, log likelihood, Akaike’s information criterion, Schwarz’s criterion) demonstrated that spatial models always outperformed conventional models. The tests based on the Lagrange Multiplier and Likelihood Ratio tests were significant at 99%. The importance of both agricultural and non-agricultural factors for determining the arable land value was confirmed. The land value increased with irrigation availability (by a mean of 2.2-fold for the set of all land uses), plot size (by 5.7% for each 1 ha increase), population size, income and location in nature reserves (11.02-12.89%).Research highlights: Results indicate the need to develop spatial models when modeling land prices by implementing GIS.