Lidar (radar laser) systems take advantage of the relatively strong interaction between laser light and aerosol/molecular species in the atmosphere. The inversion of optical atmospheric parameters is of prime concern in the fields of environmental and meteorological modelling and has been (and still is) under research study for the past four decades. Within the framework of EARLINET (European Aerosol LIdar NETwork), independent inversions of the atmospheric optical extinction and backscatter profiles (and thus, of the lidar ratio, as well) have been possible by assimilating elastic-Raman data into Ansmann et al.’s algorithm [the term “elastic-Raman” caters for the combination of one elastic lidar channel (i.e., no wavelength shift in reception) with an inelastic Raman one (i.e., wavelength shifted)]. In this work, an overview of this operative method is presented under noisy scenes along with a novel formulation of the algorithm statistical performance in terms of the retrieved-extinction mean-squared error (MSE). The statistical error due to signal detection (Poisson) is the main error source considered while systematic and operational-induced errors are neglected. In contrast to Montercarlo and error propagation formulae, often used as customary approaches in lidar error inversion assessment, the statistical approach presented here analytically quantifies the range-dependent MSE performance as a function of the estimated signal-to-noise ratio of the Raman channel, thus, becoming a straightforward general formulation of algorithm errorbars.