Time-series analysis to forecast lead prices at the London Metal Exchange (LME) to find optimal purchase periods
For a comprehensive report on objectives, methodologies and results, kindly refer to the Project Report.
For the report on historical trends of data, refer to the file "LME Lead Price - Analysis".
The project analyses the effect of various economic factors affecting lead price, and the analytics associated with deducing important trends from these factors (such as the correlation between lead stocks and prices with the Exchange Rates). Statistical and Deep Learning models such as AR, ARIMA, 1D CNN and LSTM were used to forecast lead prices using time-series modelling of historical data.
Python Version: Python 3.7
Libraries and Frameworks: Tensorflow, Keras, NumPy, Pandas, Matplotlib, Statsmodels, SkLearn, Pickle, Streamlit.