Gaussian Transformation Methods for Spatial Data

Data gaussianity is an important tool in spatial statistical modeling as well experimental data analysis. Usually field and observation deviate significantly from the normal distribution. This work presents alternative methods for transformation revisits applicability of a modified version well-known Box-Cox technique. The recently proposed method has significant advantage transforming negative sign (fluctuations) advance to positive ones. Fluctuations derived detrending cannot be transformed using common methods. Therefore, Modified technique provides reliable solution. was tested average rainfall detrended (fluctuations), groundwater level data, Total Organic Carbon wt% residuals random number generator simulating potential results. It found that competes successfully transformation. On other hand, it improved normalization or fluctuations. coding presented by means Graphical User Interface format MATLAB environment reproduction results public access.