In this contribution, we propose an alternative way to calibrate and estimate S-wave velocities using a regression analysis methodology. Prediction of S-wave velocities is critical in locations where sonic logs only have P-wave velocities available or for some reason only in discrete intervals, generally in slow formations. The proposed method applies two robust quadratic models for estimating S-wave velocities, assuming V p , clay volume and, fractional effective porosity as parameters affecting V s . Considering S-wave velocities varying according to quadratic models, the estimation process confirms previous results showing V p as the most important dependence parameter in the regressions. The preliminary results corroborate the idea that the use of the proposed quadratic model by regression analysis can represent a suitable procedure for predicting shear wave velocities in slow siliciclastic rocks.