Problem Statement: Developed a predictive model to solve the challenge of accurately pricing used cars, where values fluctuate significantly based on age, mileage, brand reputation, and technical specifications - unlike standardized new car pricing.

Technical Implementation: Engineered a Multiple Linear Regression model using Python's data science stack (Pandas, NumPy, Scikit-learn) to quantify how features like engine capacity, fuel type, and vehicle age impact resale value, achieving measurable accuracy through RMSE and R² metrics.

Data Pipeline: Processed raw automotive data by handling missing values, normalizing numerical features, and encoding categorical variables, complemented by exploratory visualizations (Matplotlib/Seaborn) to identify key pricing trends and outliers.

Deployment & Impact: Packaged the solution in an executable Jupyter Notebook with clear documentation, enabling dealerships and online marketplaces to implement data-driven pricing strategies that reduce disputes and optimize inventory turnover.

Tools and Libraries • Python • Pandas • NumPy • Matplotlib • Seaborn • Statsmodels • Scikit-learn • Jupyter Notebook

Technologies • Notebook-Based Development • One-Hot Encoding • Variance Inflation Factor (VIF) • Train-Test Split • Ordinary Least Squares (OLS) Regression • Decision Tree

Processes • Data Collection and Understanding • Data Cleaning and Preparation • Exploratory Data Analysis (EDA) • Feature Selection and Engineering • Model Building • Model Evaluation and Validation • Insight Generation