Shrimp Prices is a data analysis project focused on studying and predicting shrimp price trends between 1991 and 2021. Using historical data, it applies Polynomial Regression to model the relationship between monthly changes and market prices.
The project is built with Python, leveraging libraries such as NumPy, Pandas, Scikit-learn, and Matplotlib for data processing, statistical modeling, and visualization. It demonstrates how machine learning can be applied to real-world economic data to uncover patterns and forecast future price behavior.
Project Overview
This analysis explores shrimp price fluctuations using data collected between 1991 and 2021. The dataset includes monthly price changes, which were transformed and processed to train a polynomial regression model capable of capturing non-linear relationships between time and price variation.
The workflow includes data cleaning, feature transformation, model training and evaluation, followed by visualization of both real and predicted values to assess performance. The model’s accuracy was evaluated using Mean Squared Error (MSE).
Technical Stack
- Python – Core language for data analysis and modeling.
- NumPy & Pandas – Data manipulation and statistical computation.
- Scikit-learn – Polynomial regression and model evaluation.
- Matplotlib – Data visualization and graphical representation.
Key Learnings
- Applying polynomial regression to model non-linear relationships in real-world economic data.
- Data preprocessing and feature engineering are crucial to improving prediction accuracy.
- Visualizing data effectively helps to interpret trends and assess model performance.
- Understanding the importance of train/test splits and evaluation metrics (MSE) for reliable predictions.
- Leveraging Python libraries (NumPy, Pandas, Scikit-learn, Matplotlib) for end-to-end data analysis workflows.
This project serves as an example of applying statistical and machine learning techniques to long-term economic data, highlighting how data science can provide insights into market behavior and price prediction.