Depreciation Method Based on Information Asymmetry and Adverse Selection
Description
This dataset was developed to analyze the depreciation of electric vehicles (EVs) in the Colombian market, focusing specifically on how perceived information asymmetry affects resale values. Data collection was carried out using web scraping techniques on e-commerce platforms, primarily Mercado Libre and TuCarro, capturing EV sales information between 2019 and 2024. Key Variables: Brand: Brand of the electric vehicle (e.g., Audi, BMW, BYD). Model: Year of the vehicle model. Initial Price: Launch price of the vehicle at the time of release. Current Price: Resale price of the vehicle on e-commerce platforms. Accumulated Kilometers: Total mileage of the vehicle at the time of resale. This dataset is part of a broader research project funded by Fundación Universitaria Los Libertadores, titled "Depreciation Model for Electric Vehicles in Colombia." This project aims to develop a more accurate depreciation model that considers market dynamics, consumer perceptions, and technological advancements.
Files
Steps to reproduce
Data Collection: Platforms Used: The data was collected from e-commerce websites such as Mercado Libre and TuCarro, which list second-hand vehicles in Colombia. Timeframe: Data covers the years 2019 through 2024. Method: Web scraping techniques were applied, specifically using the Python library Selenium, to extract details like the brand, model year, initial price, current price, and accumulated kilometers of electric vehicles. Data Processing: Refinement: Raw data from the platforms was cleaned to remove irrelevant information such as vehicle color, transmission type, and fuel type, focusing only on variables essential for depreciation analysis. Target Data Construction: The cleaned data was structured into a “Target Data” set containing columns for vehicle brand, model year, initial price, current price, and accumulated kilometers. Depreciation Model Application: Baseline Method: Straight-Line Method (SLM) was first applied to the dataset as a traditional depreciation approach. Proposed Method: An adverse selection-based method, focusing on perceived information asymmetry, was developed and applied to the same dataset. This method uses accumulated mileage as a proxy for perceived quality to calculate depreciation, accounting for market perception variations. Analysis and Visualization: Error Calculation: Absolute errors were computed for both methods to evaluate accuracy. Visualization: Heatmaps and other graphical comparisons (e.g., density plots) were created to illustrate depreciation patterns over time for different vehicle brands and models. Tools and Libraries: Python Version: 3.9.0 Libraries: Selenium for web scraping, Pandas for data handling, and Matplotlib for visualizations. These steps provide a complete workflow to replicate the dataset, including data collection, processing, model application, and analysis.