Parameter Estimation

Maximum likelihood estimation, EM algorithm, residual diagnostics, and parameter transforms. See Parameter Estimation for a guide on choosing between MLE and EM.

MLE

Gradient-based optimization of the log-likelihood using JAX autodiff. Flexible: supports any differentiable parameterization via a user-defined model_fn.

EM Algorithm

Iterative variance estimation with guaranteed non-decreasing log-likelihood. Simpler setup than MLE — just pass an initial model.

Diagnostics

Tools for checking model adequacy after fitting.

Transforms

Map unconstrained parameters to positive values for variance estimation.