Estimation of density and other elastic parameters is important in characterizing conventional, unconventional and carbon dioxide (CO2) sequestrated reservoirs. While it is possible to estimate velocities from the conventional P-wave seismic data, it is difficult to accurately estimate density from such data. Because the converted wave (P-SV) reflections are more sensitive to density compared to the primary (PP) reflections, using both PP and P-SV data in a multicomponent inversion has been shown to accurately estimate density compared to using P-wave data alone. The problem of such multicomponent inversion is, however, nonlinear, requires simultaneous optimization of multiple sets of data and has multiple optimal solutions. Such simultaneous optimization is better handled by a nondominated sorting genetic algorithm in comparison to the classical approach. Applying the nondominated sorting genetic algorithm to multicomponent synthetic seismic data here, we demonstrate that this method can potentially extract the subsurface elastic parameters and density.