Modern polarimetric weather radars typically provide reflectivity, differential reflectivity, and specific differential phase shift, which are used in algorithms to estimate the parameters of the rain drop size distribution (DSD), the mean drop shape, and rainfall rate. A new method is presented to minimize the parameterization error using the Rayleigh scattering limit relations multiplied with a rational polynomial function of reflectivity-weighted raindrop diameter to approximate the Mie character of scattering. A statistical relation between the shape parameter of the DSD with the median volume diameter of raindrops is derived by exploiting long-term disdrometer observations. On the basis of this relation, new optimal estimators of rain microphysical parameters and rainfall rate are developed for a wide range of rain DSDs and air temperatures using X-band scattering simulations of polarimetric radar observables. Parameterizations of radar specific path attenuation and backscattering phase shift are also developed, which do not depend on this relation. The methodology can, in principle, be applied to other weather radar frequencies. A numerical sensitivity analysis shows that calibration bias and measurement noise in radar measurements are critical factors for the total error in parameters estimation, despite the low parameterization error (less than 5%). However, for the usual errors of radar calibration and measurement noise (of the order of 1 dB, 0.2 dB, and 0.3 $\hbox{deg}\ \hbox{km}^{-1}$ for reflectivity, differential reflectivity, and specific differential propagation phase shift, respectively), the new parameterizations provide a reliable estimation of rain parameters (typically less than 20% error).