Anthonykennetho/maize-yield-vs-rainfall-regression-jos

A linear regression project exploring the relationship between rainfall and maize yield in Jos, Nigeria, using synthetic data and exploratory data analysis. # Maize Yield Prediction This project uses **simple linear regression** to predict **maize yield** (bags per hectare) based on **rainfall** (mm). The goal is to understand the relationship between these two variables and make predictions. ## Dataset The dataset includes the following columns: - **Year**: The year of observation. - **Month**: The month of observation. - **Rainfall_mm**: Amount of rainfall (mm). - **Maize_Yield_bags_per_ha**: Maize yield (bags per hectare). ## Model A **simple linear regression** model is used to predict **maize yield** based on **rainfall**. ### Key Metrics: - **Pearson Correlation Coefficient**: **0.808** Indicates a strong positive linear relationship between **rainfall** and **maize yield**. - **R-squared (R²)**: **0.658** About **65.8%** of the variance in **maize yield** is explained by **rainfall**. - **Mean Absolute Error (MAE)**: **2.01 bags/ha** On average, the model's predictions are off by about **2 bags per hectare**. - **Mean Squared Error (MSE)**: **6.29** This indicates moderate error, penalizing larger errors more heavily. ## Conclusion The model provides a solid baseline for predicting **maize yield** based on **rainfall**. While the R² value suggests a moderate relationship, approximately **34.2%** of the variance in maize yield is unexplained. This indicates that other factors (e.g., **seasonality**, **soil type**, **temperature**) could further improve the model. ## Future Improvements - **Feature Engineering**: Add seasonal features like **month**, **year**, **soil types** and **temperature** to capture temporal trends. - **Advanced Models**: Explore **polynomial regression** or machine learning models (e.g., **random forests**, **gradient boosting**) to capture non-linear relationships. - **Validation**: Test the model on a validation or test set to ensure it generalizes well. ## Getting Started ### Requirements - Python 3.x - Required libraries: - pandas - numpy - scikit-learn - matplotlib