Regression Analysis of Energy Consumption and Spatial Needs

This project analyzes square footage needs by examining a variety of influencing factors including number of occupants, average temperature, energy consumption, building type, and day of the week. Using both Scikit-learn's LinearRegression and Statsmodels' OLS, we aim to uncover insights into how these variables affect spatial dimensions. Key analysis stages include data exploration, preprocessing with one-hot encoding, regression modeling, and model evaluation. The final model achieves an R² of 0.844, indicating strong predictive accuracy. Additional experiments using a single feature (energy consumption) yielded R² = 0.524, supporting its significant role. This project highlights the importance of multi-feature regression, encoding categorical data, and interpreting statistical outputs for data-driven decision-making in energy and space planning. Dataset: test_energy_data.csv Tools: Python, Pandas, Numpy, Matplotlib, Scikit-Learn, and Statsmodels