vcov_msfa: Variance matrix of MLE estimates for a MSFA model
In rdevito/MSFA: Maximum Likelihood Estimation and Bayesian analysis of MSFA models

Description Usage Arguments Details Value

Computes the inverse observed information for a MSFA model

1	vcov_msfa(X_s, mle, getgrad = TRUE)

`X_s`	List of lenght S, corresponding to number of different studies considered. Each element of the list contains a data matrix, with the same number of columns P for all the studies.
`mle`	The object returned by `ecm_msfa`.
`getgrad`	Should the function return also the gradient at `mle`? Default is `FALSE`.

Numerical differentiation is employed to obtain the observed information matrix at a given parameter values, so that when the parameter values equals the MLE the function returns the estimated variance matrix of the fitted model. The method is rather inefficient, and it may lead to long computations, though the function is designed to be called only once after the estimation has been carried out. However, it would be relatively straightforward to employ analytical differentiation at least for the log-likelihood gradient, and this may be implemented in future releases of the code.

A list with exactly the same structure of the three slots Phi, Lambda_s and Psi_s of mle, but containing the standard errors rather than the point estimates. Furthemore, slots for the hessian matrix and the gradient at mle are included, the latter is not NULL when getgrad is TRUE.