In order to develop more potent, selective and less toxic steroidal aromatase (AR) inhibitors, molecular docking, 2D and 3D hybrid quantitative structure–activity relationship (QSAR) study have been conducted using topological, molecular shape, spatial, structural and thermodynamic descriptors on 32 steroidal compounds. The molecular docking study shows that one or more hydrogen bonds with MET374 are one of the essential requirements for the optimum binding of ligands. The QSAR model obtained indicates that the aromatase inhibitory activity can be enhanced by increasing SIC, SC_3_C, Jurs_WNSA_1, Jurs_WPSA_1 and decreasing CDOCKER interaction energy ( E CD), IAC_Total and Shadow_XZfrac. The predicted results shows that this model has a comparatively good predictive power which can be used in prediction of activity of new steroidal aromatase inhibitors.