Background: The agricultural market is distinct from the traditional stock market. Agricultural contracts typically have fixed expiration months that align with production and harvest cycles. This results in a more structured yet less liquid market compared to the continuous trading seen in stocks.

Project Objectives:
- Baseline Model Development: We will start by building a baseline predictive model using traditional market data such as historical prices, volumes, and contract specifications. Models like Linear Regression, ARIMA, or Gradient Boosting Machines (GBM) will serve as our starting points, allowing us to understand the core dynamics of these markets.
- Incorporation of Alternative Data: Once we establish a reliable baseline, the project will explore enhancing the predictive performance by incorporating alternative data sources:
-- Weather Data: Impact of temperature, precipitation, and extreme weather events on crop yields and livestock conditions.
-- News Sentiment: Analysis of market sentiment from news articles and reports.
-- FX Market Data: Influence of currency fluctuations, especially since many commodities are priced in U.S. dollars.
-- Satellite Imagery: Assessment of crop health and acreage through remote sensing data.
-- Other Sources: Any additional relevant datasets that can provide deeper insights into market fundamentals.