Abstract
The multiple regression model with parameters of N-value and depth implies a form of shear wave velocity with a specific physical meaning, but these two parameters are not independent of each other. The research introduced statistical analysis techniques of the stepwise selection, the "rule of thumb" test, and the "RSM" test to overcome the multicollinearity problem. By using 491 data collected from the Taipei basin, the results show that only SM soil failed to pass the criteria while the CL and ML soils successfully did. Thus, the multiple regression model in a form of Vs=aNbDc may be used in some soils cases.
