Hi Rick,   Thank you for your suggestions and resources.   The response (A1, A2) for my data is continuous longitudinal data and analogous to eqn. (1)-(4), concatenating response vectors of multiple features for individual i results in :  We can assume all the random variables in eqn.(5) follow a joint multivariate normal distribution( ) to build the multivariate nonlinear mixed model.  thus  Based on eqn. (7) we can get the log likelihood function. However, how to write the variance in eqn. (7). Could you give me some suggestions? Thanks.         ... View more