Linear Regression: Advanced Modeling
This project aimed at developing a real estate evaluation system using linear regression methodology, which is one of the machine learning techniques. The dataset is a random sample of 5000 properties available for sale in the city of Rio de Janeiro, Brazil. The data includes the following features: Value (R$), Property Area (m²), Distance from the property to the beach (km) (in a straight line), and Distance from the property to the nearest pharmacy (km) (in a straight line). Due to the skewed data distribution, a logarithmic transformation is necessary to achieve satisfactory results with the linear regression method.
The project aims to explore advanced linear regression modeling techniques to accurately predict real estate values based on the given dataset.
The project focuses on the following:
- Apply transformations before training models
- Implement regressions using Statsmodel and Sklearn
- Learn advanced modeling techniques
- Obtain point predictions
- Interpret estimated coefficients
- Perform graphical analysis of the results