Skip to content

AutoMMM v0.1.5

Latest
Compare
Choose a tag to compare
@nialloulton nialloulton released this 22 Apr 13:49
· 6 commits to main since this release
a757a05

AutoMMM v0.1.5 Release Notes

AutoMMM is an experimental open-source automodeller for feature selection in marketing mix models, designed to automate the process of building base marketing mix models and feature selection methods using Ridge Regression and the GPT-4 AI model. In this release, we've introduced a new Vector Autoregression (VAR) model feature, along with improvements to existing features.

New Features

  • VARModel: The new VARModel class allows users to build, fit, and analyze Vector Autoregression (VAR) models for time series data with support for exogenous variables and impulse response functions.

Improvements

  • Updated existing features for better performance and usability.
  • Minor bug fixes and stability enhancements.

Installation

To install AutoMMM v0.1.5, run the following command:

pip install autoMMM==0.1.5