A study on elastic moduli identification method of 2.5 dimensional C/CSi composite with uncertainty is conducted in this paper. The uncertain parameters to be identified are divided into the summation of mean value and a deviation term based on which the iteration formulations for the uncertainty identification are derived. The initial equivalent elastic modulus of the composite is predicted by using of stiffness average method. E11, E22 and G12 are selected as the identifying parameters through relative sensitivity analysis, after deterministic identification, a reference model for study of uncertainty identification method is obtained. Based on this reference model, adopting the Latin Hypercube sampling method to construct random test samples, then the uncertain elastic parameters identification method is conducted based on stochastic model updating method. Numerical results show that When considering the uncertainty in elastic modulus of 2.5 dimensional C/SiC composite, the uncertain parameter identification method proposed can be used for accurately identifying the mean value and deviation of the elastic modulus, the dynamical finite element model with statistical significance can be constructed at the same time.