The aim of this paper is to provide an introduction of new imputation algorithms for estimating  missing  values  from official statistics in larger data sets of data pre-processing , or outliers. The goal is to  propose  a new algorithm called IRMI (iterative robust model-based imputation).This algorithm is able to deal with all challenges like representative and non-representative outliers and a mixture of different distributions of variables. This algorithm is compared to the algorithm IVEWARE to illuminate the advantages and disadvantages of different techniques for imputation in artificial data and real data sets from official statistics , with respect to robustness are proposed, especially in presence of outliers the model-based of new algorithm  is preferable.