Abstract: The identification of the joint structural parameters of a machine tool is one of important elements for improved and efficient design. In this paper, a new method for combining the time series analysis with finite element model, by using condensation technique, is presented to identify these joint structural parameters. First, the modal, parameters of the structure are extracted from a similar machine or simplified prototype model by both experiments and analysis using Autoregressive Moving Average Vector (ARMAV) models. Next, the finite element model of the structure is reduced to a smaller degrees of freedom using a new condensation technique to match with the order of the ARMAV models. For this purpose, certain complex frequencies obtained from the ARMAV models are used to formulate the condensed mass and damping matrices. Finally, by equating the modal parameters from the ARMAV models with those of the condensed system, the unknown joint stiffness and damping coefficients are identified. The simulation study of theoretical case is used to check the sensitivity of this new.condensation method. Experiment is also carried out to implement this methodology to determine the joint stiffness and damping coefficients for a column model. The results show that the presented new method has sufficient accuracy for the engineering purpose.