LSTM Oil Price Time Series Forecasting and Performance Metrics Evaluation Python code to perform oil price forecasting. This is an univariate LSTM model and then perform metrics evaluation
Package Version
python 3.6.8
pandas 0.24.1
Keras 2.2.4\
This python demonstrates the univariate oil price forecasting using LSTM deep learning model
The sliding window is used to creating training and validation in the time series data
The standard scalers provided by Sklearn is used.
Both the time the forecasted and actual oil prices are plotted to visualize.\
Then using sklearn.metrics and math various metrics such as
- Mean Absolute Error
- Mean Squared Error
- Root Mean Square Error
- R2 Score
- Adjusted R2
are computed on predicted data.
For multivariate oil data is present at https://github.com/siddhaling/Oil-Price-WTI-Database-For-Prediction-and-Forecasting
Research is published in
https://www.sciencedirect.com/science/article/abs/pii/S0360544221012111
Cite as: Siddhaling Urolagin, Nikhil Sharma, Tapan Kumar Datta, A combined architecture of multivariate LSTM with Mahalanobis and Z-Score transformations for oil price forecasting, Energy, Volume 231, 2021, 120963, ISSN 0360-5442, https://doi.org/10.1016/j.energy.2021.120963.
https://medium.com/@dr.siddhaling
Dr. Siddhaling Urolagin,
PhD, Post-Doc, Machine Learning and Data Science Expert,
Passionate Researcher, Focus on Deep Learning and its applications,
dr.siddhaling@gmail.com