TSPred Package for R : Functions for Baseline-Based Time Series Prediction

TSPred Package for R : Functions for Baseline-Based Time Series Prediction

Authors: Rebecca Pontes Salles (rebeccapsalles@acm.org) and Eduardo Ogasawara (eogasawara@ieee.org)

Description: Functions for time series prediction and accuracy assessment using automatic ARIMA modelling. The generated ARIMA models and its yielded prediction errors are intended to be used as baseline for evaluating the practical value of other time series prediction methods and creating a demand for the refinement of such methods. For this purpose, benchmark data from prediction competitions may be used.

Available at CRAN: http://cran.r-project.org/web/packages/TSPred/index.html

Tutorial: https://sourceforge.net/p/gpca/wiki/TSPred/

Reference manual: TSPred.pdf

Santa Fe Competition Dataset

EUNITE Competition Dataset

CATS Competition Dataset

NN3 & NN5 Competition Dataset

Acknowledgements: The authors thank CNPq for partially sponsoring this work.

Comments are closed.