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.