The CRD, founded on the Basel recommendations, put the whole risk assessment of the banks on new bases, but considering that the data series available for many credit institutions are not old and accurate enough for carrying out consistent estimations, as well as the quantity of the default data is not adequate in many cases, therefore calculating the own LGD values comes up against numerous difficulties. Among others in the Hungarian practice the banks are not able to take advantage of the theoretical opportunities given by the Basel rules in many cases, because the necessary conditions of the secondary market of loans and bonds do not exist. For that very reason the credit institutions have to focus on the historic collecting of data and on preparing the most possible accurate predictive models for the sake of exactly quantifying the credit risk. The aim of this research was to study the characteristic features of the retail mortgage loan's LGD and to prepare a model for quantifying it more exactly and accurately. The present analysis should be useful to professionals in Credit Risk Analysis field, especially to those who are concerned with the Eastern-European counties.