Abstract A separable covariance model can describe the among-row and among-column correlations of a random matrix permits likelihood-based inference with very small sample size. However, if assumption separability is not met, data analysis may misrepresent important dependence patterns in data. As compromise between unstructured estimation, we decompose into component complementary ‘core’ matri...

Abstract. Stochastic dynamic analysis of linear or nonlinear multi-degree-of-freedom systems excited by multi-variated processes is usually conducted using digital Monte Carlo (MC) simulation. Since in structural few modal shapes contribute to the response nodal space, computational burden MC simulation mainly related input process. Usually, generation samples Gaussian process performed with ai...