econometron.utils.estimation.results

• compute_stats function from econometron.utils.estimation.results

Overview

This module provides advanced tools for post-estimation inference, diagnostics, and reporting in econometric and Bayesian workflows. It includes utilities for computing standard errors and p-values via numerical Hessians.

Theory

Statistical inference after model estimation is crucial for understanding parameter uncertainty and model fit. This module supports:

• Numerical Hessian-based inference: Standard errors and p-values are computed by numerically approximating the Hessian of the objective function, inverting it to obtain the covariance matrix, and applying normal theory for inference.

Functions

Function	Description
compute_stats	Computes standard errors and p-values for parameter estimates using a numerical Hessian

Usage

• Use compute_stats after optimization to obtain standard errors and p-values for parameter estimates.