Abstract:In this paper, the optimal estimation method of rock shear strength parameters is studied. Aiming at the heteroscedasticity problem in rock shear strength data, a weighted linear regression method is introduced. In the linear regression analysis of rock shear strength, it is generally assumed that the observation errors of dependent variables obey the same normal distribution, but the heteroscedasticity problem in triaxial strength data is ignored. In this paper, the open-pit sandstone in the southwest of Sichuan Province is used as the research object, and the triaxial strength data of sandstone under different confining pressures are obtained. The weighted linear regression method is used to analyze and solve the problem of heteroscedasticity of the data. This paper first introduces the theoretical basis and related statistical methods of rock shear strength, and then elaborates the concept and application scope of weighted linear regression. The theoretical formula and test method of weighted linear regression are proposed, and the definition and calculation method of weight in weighted regression are explained. Finally, the experimental data of open-pit sandstone in the southwest of Sichuan Province are verified. The research results prove the effectiveness of the weighted least squares method in the estimation of rock shear strength parameters. The research results can provide relevant theoretical basis for the selection of shear strength parameters.