We propose a methodology for anonymization of records by aggregating sequential records. This methodology is applied to real valued, binary, nominal and ordinal data. We propose two metrics to examine the properties of the proposed method. Empirical results on simulated dataset, confirms the utility of the proposed approach.