Compute case-wise and cluster-wise derivative for mixed effects models with respect to fixed effects parameter, random effect (co)variances, and residual variance.
merDeriv is a free, open source R package for computing derivatives of mixed models estimated via lme4. Allows users to compute scores (casewise first derivatives of likelihood function) and variance-covariance matrix of all estimated parameters, including random effect (co)variances.
For further information, see:
Wang, T. & Merkle, E. C. (2018). Derivative Computations and Robust Standard Errors for Linear Mixed Effects Models in lme4.. Journal of Statistical Software, 16(9), 1–16.