An Improved GWR Approach for Exploring the Anisotropic Influence of Ore-Controlling Factors on Mineralization in 3D Space

Spatial non-stationarity is a common geological phenomenon, and the formation of orebodies typical non-stationary process. Therefore, quantitative study relationships between mineralization its controlling factors in 3D space great significance for metallogenic prediction. Geographically weighted regression (GWR) an effective method exploring spatial by measuring nearness data. However, affected not only distances but also related to direction. Traditional GWR cannot address that arises from differences To this issue, we propose improved characterize directional characteristics introducing direction weight GWR. The anisotropic influence can be obtained comparing performance models with different weights on terms. A case Xiadian Dayingezhuang gold deposits, Jiaodong Peninsula, Eastern China was carried out verify nature ore-controlling factors. First, multi-collinearity OLS (ordinary least squares) diagnosis variables were performed necessity exploration demonstrated. Second, applied explore among global R2 value OLS, evaluating local R2, values testing t-statistic values, analyzing autocorrelation residuals calculating stationary index parameter estimates explanatory variables. Third, applied, relationship analyzed. Finally, validated model bandwidths kernels, importance fault zone further verified.