Predictive Modeling of Automobile Prices Using Feature Engineering and Regression Techniques

Over the years, Accessories for automobile have become necessities to be used in personal vehicles for commuting from home to office as well as traveling during vacations. Buying a new or used automobile is a decision that has to made with caution, especially since one of the hardest things to do is to sell an old automobile when the time comes to do so. Due to the high rate of new vehicle automobile with reducing purchasing power, lots of buyers find it increasingly harder to choose if they are better off purchasing a new automobile or ending up choosing a used. We have developed various methods to predict the price of automobile vehicles as per the market trends to avoid this situation. Therefore, we propose a price prediction model that helps both buyers and sellers make the right decision for their business and personal needs. The model employs a type of machine learning technique called regression in order to achieve higher accuracy in predicting results. Using Recursive Feature Elimination (RFE) and Variance Inflation Factor (VIF) techniques to determine the most important contributors to automobile prices, we use Ordinary Least Squares (OLS) regression to optimize the model. Despite its relative simplicity, the study shows that this technique is both effective and efficient, producing accurate predictions helpful to both sellers hoping to get the best price and buyers in search of a fair price. It is shown that the model proposed here outperforms other available methods, which contributes to increases in the preciseness of pricing in the automotive market.