Regularized Maximum Likelihood Estimation for the Random Coefficients Model in Python

Abstract

We present PyRMLE (Python regularized maximum likelihood estimation), a Python module that implements regularized maximum likelihood estimation for the analysis of Random coefficient models. PyRMLE is simple to use and readily works with data formats that are typical to Random coefficient problems. The module makes use of Python’s scientific libraries NumPy and SciPy for computational efficiency. The main implementation of the algorithm is executed purely in Python code, which takes advantage of Python’s high-level features. The module has been applied successfully in numerical experiments and real data applications. We demonstrate an application of the package in consumer demand.

Related Papers

• → Using Python for scientific computing: Efficient and flexible evaluation of the statistical characteristics of functions with multivariate random inputs(2012)20 cited
• → PySSM: APythonModule for Bayesian Inference of Linear Gaussian State Space Models(2014)6 cited
• → Numerical Computing in Python(2004)4 cited
• → Data for: PyFEST - a Python code for accurate frequency estimation(2020)3 cited
• → Nonparametric Estimation of the Random Coefficients Model in Python(2021)