Backpropagation-Neural-Network-for-Multivariate-Time-Series-Forecasting

This is a project about Backpropagation Neural Network for Multivariate Time Series Forecasting (multi-input single-output: 2 inputs and 1 output)

There are several steps in multivariate time series forecasting using the Backpropragation Neural Network. Here are the steps to take:

1. Pre-processing (Min Max Normalization)
2. Initialize Network (Inisialisasi Bobot)
3. Feed Forward Propagation
4. Backpropagation
5. Train (use execution time)
6. Predict
7. Forecast Result
8. Forecast Errors
9. Accuracy Result (MAE, MSE, RMSE, MAPE RESULT)

This source code is made in Python 3.0 using Anaconda Jupyter.

For any suggestions and discussion, please reach out to my Linkedin or email (only): zahra.informatika@gmail.com.

Thank you.

Regards, Zahra