Estimation of housing prices in Ecuador using hedonic and geostatistical models: A comparative analysis

This paper compares the results of two models, one hedonic and other geostatistical, when obtaining housing price estimates in Rumiñahui canton, Ecuador. In first model, different parameterizations variables that make up model are used to obtain best predictions. For geostatistical case, predictions composed a trend function depends on certain characteristics spatial error term, which is modeled from residual variogram. The performance each compared by an analysis prediction errors validation data set. indicate better for since it considers, addition inherent house, effect its location selling