This is the official repository of the research poster Clustering Subsecuencial de Series de Tiempo: Evidencia de Patrones Temporales en el tipo de cambio USD/MXN, originally presented in its spanish version at the National Seminar Escuela de Probabilidad y Estadística 2020 (EPE2020). Its an academic event organized by CIMAT (A government-funded research institute, internationally renowned, which main focus is Mathematical Research).
There are three types of content in this repository.
- Data scraping, preparation and clustering: Python
- Data analysis, synthesis and presentation: R
- Tables, Figures, Formulas within the beamer poster: LaTeX
In order to conduct a wide list of experiments variations with all candidate trigger events searching within all the historical prices, it was necessary to perform a parallelization of the project, mainly in the process where the MASS algorithm was used to search for subsequential patterns.
This work was aimed at testing the capabilities of the Mueen's Algorithm for Similarity Search (MASS), we acknowledge the work of the original authors which can be consulted for further technical details.
The python implementation that was utilized for this work is the MASS-TS python package, which can be found and installed at https://pypi.org/project/mass-ts/
- Clone repository
Clone entire github project
git@github.com:IFFranciscoME/STSC-Temporal-Patterns.git
(optional) create a virtual environment
virtualenv venv
(optional) activate virtual environment
source ~/venv/bin/activate
and then install dependencies
pip install -r requirements.txt
J.Francisco Munnoz - IFFranciscoME - Is an Associate Professor in the Mathematics and Physics Department, at ITESO University.
GNU General Public License v3.0
Permissions of this strong copyleft license are conditioned on making available complete source code of licensed works and modifications, which include larger works using a licensed work, under the same license. Copyright and license notices must be preserved. Contributors provide an express grant of patent rights.
Contact: For more information in reggards of this project, please contact francisco.me@iteso.mx