Alcohol Sales Forecasting Using LSTM

This project implements a Long Short-Term Memory (LSTM) neural network to forecast monthly alcohol sales in the United States. The dataset is obtained from the Federal Reserve Economic Data (FRED). The entire workflow, including preprocessing, model training, evaluation, and visualization, is performed using PyTorch and standard data science libraries in Python.

Project Overview

• Objective: To forecast future alcohol sales using historical monthly data.
• Dataset: Monthly Alcohol Sales (FRED ID: S4248SM144NCEN).
• Model: LSTM-based time series forecasting model.
• Tools Used: PyTorch, Pandas, Matplotlib, NumPy, Scikit-learn.

Repository Structure