メッシュデータ解析の一方法としての空間相関分析法の提案 : その3 空間影響関数モデルの有効性と問題点

概要

論文の詳細を見る
The aim of this paper is to examine the applicability of the spatial influence model which has been theoretically proposed in the previous paper. At first, the criteria which measure the efficiency of the mathematical models on the urban planning are discussed and three measures are defined : 1) the correlation coefficient between the original data and the estimated values, 2) the number of parameters in the model, 3) Akaike's information criterion AIC. Nextly, the spatial influence function model (SIF) and the ordinary multi-regression model (MR) are adapted on the same mesh-data of the land use. The results show that SIF model is better than the MR model although MR is simple as compared with SIF. However the problem which is a variation of the multi-colinearlity is observed in the calculating process of the parameters of SIF model. These results show that SIF model is efficient on the estimation or prediction of mesh-data, however, it must be avoid to give any meaning on the values of estimated parameters from a view point of planning.
1987-07-30