본문 바로가기
  • Home

Estimating Farmland Prices Using a Regression-Kriging Model

Son, Hak-gi 1 Lee, Changro 2

1농촌경제연구원
2서울대학교

Accredited

ABSTRACT

This study estimates farmland prices in Kimje City and analyzes the result empirically. Previous studies tend to utilize the income approach more often than not, which has limits in reflecting site-specific characteristics. Thus, we employ the market approach instead. The input data are prices of standard lots estimated by property valuation experts. We apply the market approach to the data and make the first use of the regression-kriging model which has advantage of dealing with both attribute and location information of farmland. When inspecting the model performance based on the test data, we confirmed that the price accuracy from the regression-kriging model was improved substantially compared with that of the usual regression model. That is, the estimated price became more accurate through spatial dependence being incorporated in the model building process. We also could avoid the subjectivity in quantifying spatial relationship using a kriging technique. This improvement is also confirmed visually through reviewing the spatial distribution of estimated-to-actual price ratios.

Citation status

* References for papers published after 2023 are currently being built.