close
Skip to content

Latest commit

 

History

3 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

TEAM (The Time Series Exhaustive Automatic Modeling) team logo

TEAM is a software designed to automatically identify the best REG-ARIMA models for performing seasonal and calendar adjustments.

Overview

It was developed with the following objectives:

  • Reduce the time needed to select models through a tool that automates the model identification phase.
  • Automatically identify a set of suitable models.
  • Address TRAMO issues by offering an automatic identification tool with a robust methodology that allows obtaining optimal models.

To achieve these objectives, it is based on the following basic principles:

  • Exhaustive model search: all possible models are tested (within the combinations defined by the user).
  • Provide alternative models: TEAM offers multiple models (5 by default) instead of a single model.
  • User-adjustable: the user can specify the models to test.

TEAM leverages the JDemetra+ algorithms and makes use of JDemetra R packages, integrating proven seasonal adjustment tools within its automated workflow.

JDemetra+ is a time series analysis software, whose core routines are written in Java and also accessible via a Graphical User Interface (GUI).

For more information about JDemetra+ see: Documentation and Blog

Meanwhile, rjdverse contains all the information of the R packages focused on recreating JDemetra+ algorithms.

Methodology

The methodology used by TEAM for automatic model detection can be consulted at the following link, which provides a reference to understand the procedure, the selection criteria, and the stages of the modeling process.

Methodology

Installation

Installation instructions can be found in the link below.

Installation Guide

Getting Started with TEAM

Basic instructions on how to use TEAM can be found in the link below. It includes the way TEAM is called and a basic example. It also features a short annex with auxiliary functions to help you use the software.

Basic Guide

For more information about exhaustive outlier detection you could see:

Quarterly Case | Monthly Case

WIP and Disclaimer

Work in Progress

This project is currently under active development.

Disclaimer

This repository and its contents are provided for informational and technical purposes only.
Use of this code is at your own risk.

The Instituto Nacional de Estadística (Spanish Statistical Office) provides this software "as is", without warranty of any kind, either expressed or implied.

By using this repository, you acknowledge that:

  • You are responsible for reviewing and testing the code before using it in any production or critical environment.
  • The Instituto Nacional de Estadística is not liable for any direct, indirect, or consequential damages arising from the use of this software.

Package Maintenance and contributing

Any contribution is welcome and should be done through pull requests and/or issues. pull requests should include updated tests and updated documentation. If functionality is changed, docstrings should be added or updated.

Licensing

The code of this project is licensed under the European Union Public Licence (EUPL).

About

Exhaustive automatic modelling of economic time series

Resources

Stars

2 stars

Watchers

0 watching

Forks

Releases

Packages

Used by

Contributors

Languages