Twitch Python Finance - ARMA Modelling with Arbitrary/Non-Consecutive Lags and Exogenous Variables in Python using Sympy: From Matrix Mathematics and Statistics to Parallel Computing: Part 1

Presenter: Jonathan Legrand from Refinitiv

Length: 51 mins

This article constructs a Python function allowing anyone to discern all Maximum Likelihood Estimates of any ARMA(p,q) model using matrices, using the Python library 'Sympy'. We go through the fundamentals of ARMA models, the Log-Likelihood estimation method, matrix constructions in Sympy, and then put it all together to build our generalised model. In this 1st Part/Draft, we lay the foundations needed before continuing and make one function to encapsulate any ARMA model, with or without arbitrary lags.

Full article and details about the steps to retrieve the data can be found in the following article.

Related Links