The aim of the Crude Oil Price Forecasting project is to develop accurate and reliable models for predicting future crude oil prices, providing valuable insights for decision-making in areas such as energy policy and investment strategies.
Here are some key points about the project:
-
Dataset: The project utilizes a dataset that includes historical data on crude oil prices, typically starting from the year 1987. Additionally, the dataset may contain other economic and market indicators that have a significant impact on crude oil prices.
-
Forecasting Techniques: The project employs a range of time series forecasting techniques to predict crude oil prices. These techniques may include Autoregressive (AR), Moving Average (MA), Autoregressive Integrated Moving Average (ARIMA), Seasonal Autoregressive Integrated Moving Average (SARIMA), Long Short-Term Memory (LSTM), and other suitable models.
-
Applications: The project has practical applications for various stakeholders such as investors, traders, and policymakers.
Developed and Deployed a web app that analyzes past data and Future forecasted values of crude oil using Machine Learning models. Check out the app here: https://suraj4502-forecasting-project.streamlit.app/
◉Train Mean Absolute Error: 0.9293671837554082
◉Train Root Mean Squared Error: 1.323646028008785
◉Test Mean Absolute Error: 1.4509820900007175
◉Test Root Mean Squared Error: 1.9774892106388844
The following Python libraries have been used to build this project.
matplotlib==3.6.3
numpy==1.23.2
pandas==1.4.3
Pillow==10.0.0
plotly==5.13.0
prophet==1.1.4
scikit_learn==1.3.0
seaborn==0.12.2
streamlit
tensorflow==2.12.0
xlrd
