MTSCNeuralForecast:Python and R scripts for time series forecasting, clustering, anomaly detection and imputing missing values.
time series forecasting, clustering, anomaly detection and imputing missing values. It is a combination of R and Python scripts.
The clustering (R and Python) , anomaly detection and missing value imputation is done in R while the forecasting is done in Python.
If you find these scripts useful or if you use any of the script in your research or work please consider citing it as follows:
@article{tadayon2020clustering,
title={A clustering approach to time series forecasting using neural networks: A comparative study on distance-based vs. feature-based clustering methods},
author={Tadayon, Manie and Iwashita, Yumi},
journal={arXiv preprint arXiv:2001.09547},
year={2020}
}
-
It handles anomaly detection using Twitter Anomaly detection.
-
It handles multiple techniques for imputing missing values.
-
It handles distance based clustering (Dynamic Time Warping) and feature based clusterings.
-
It supports multivariate time series forecasting using static and time series data.
- You need to have numpy, pandas, scikit-learn, tensorflow and Keras. The R libraries were mentioned in the scripts and can be installed using install.packages command.
- This software is released under the Apache 2.0 License.
- If you have any question or concern about this package please feel free to reach out to me at [email protected]