{"id":18351948,"url":"https://github.com/nickenshidqia/predict_price_of_used_car_using_linear_regression","last_synced_at":"2025-04-10T00:37:37.196Z","repository":{"id":213125097,"uuid":"733095009","full_name":"nickenshidqia/Predict_Price_of_Used_Car_Using_Linear_Regression","owner":"nickenshidqia","description":"Build a machine learning model that can predict the price of used car","archived":false,"fork":false,"pushed_at":"2023-12-18T15:42:11.000Z","size":1998,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-10T00:37:33.439Z","etag":null,"topics":["car-price-prediction","data-science","linear-regression","machine-learning","prediction"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/nickenshidqia.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2023-12-18T14:52:35.000Z","updated_at":"2023-12-18T15:43:57.000Z","dependencies_parsed_at":"2023-12-18T17:06:30.553Z","dependency_job_id":null,"html_url":"https://github.com/nickenshidqia/Predict_Price_of_Used_Car_Using_Linear_Regression","commit_stats":null,"previous_names":["nickenshidqia/predict_price_of_used_car_using_linear_regression"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nickenshidqia%2FPredict_Price_of_Used_Car_Using_Linear_Regression","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nickenshidqia%2FPredict_Price_of_Used_Car_Using_Linear_Regression/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nickenshidqia%2FPredict_Price_of_Used_Car_Using_Linear_Regression/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nickenshidqia%2FPredict_Price_of_Used_Car_Using_Linear_Regression/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/nickenshidqia","download_url":"https://codeload.github.com/nickenshidqia/Predict_Price_of_Used_Car_Using_Linear_Regression/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248138007,"owners_count":21053775,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["car-price-prediction","data-science","linear-regression","machine-learning","prediction"],"created_at":"2024-11-05T21:33:47.986Z","updated_at":"2025-04-10T00:37:37.182Z","avatar_url":"https://github.com/nickenshidqia.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Machine Learning Project Using Linear Regression to Predict Price of Used Car\r\n\r\n## Project Description\r\n\r\n**Problem :**  \r\nBuying or selling a used car can be a complex process, and determining a fair market value for a used car is often subjective. This project addresses the challenge of predicting the price of a used car based on its specifications.\r\n\r\n**Challenges :**  \r\nBuild a machine learning model that can predict the price of used car\r\n\r\n## Project Goal\r\n\r\nBy leveraging machine learning techniques and historical data, the goal is to develop a model that provides accurate and reliable estimates of a used car's market value, taking into account various features and attributes.\r\n\r\n## Tools \u0026 Library Used\r\n\r\n[\u003cimg src=\"./image/python-logo-2.png\" alt=\"python-logo\" width=\"50\"/\u003e](https://www.python.org/) \u0026nbsp;\r\n[\u003cimg src=\"./image/jupyter-logo.png\" alt=\"jupyter-logo\" width=\"50\"/\u003e](https://jupyter.org/) \u0026nbsp;\r\n\r\n## Project Result\r\n\r\n[Click here to get full code](https://github.com/nickenshidqia/Predict_Price_of_Used_Car_Using_Linear_Regression/blob/648fc30d7b33b1878e4afb321b29416d5d5ccf30/Used%20car%20price%20prediction.ipynb)\r\n\r\n### Dataset\r\n\r\n\u003cimg src=\"./image/data_car.jpg\" alt=\"\" width = \"800\"/\u003e    \r\n \r\n- Brand ==\u003e BMW is generally more expensive than Toyota\r\n- Mileage ==\u003e the more car is driven, the cheaper it should be\r\n- EngineV ==\u003e sports car have larger engines than economy cars\r\n- Year of production ==\u003e the older the car, the cheaper it is, with exception of vintage vehicles  \r\n \r\n### Exploring the PDF's (Probability Density Function)  \r\n\u003cimg src=\"./image/bar_car.png\" alt=\"\" width = \"600\"/\u003e  \r\n  \r\n**Price**  \r\n- this is dataset about used car, therefore $300,000 is an excessive price\r\n- Ouliers are great issue for OLS ==\u003e remove top 1% of distributions !!  \r\n \r\n**Engine Volume**  \r\n- The situation with engine volume is very strange\r\n- From google search, Car engine volumes are usually below 6.5l, so we're gonna select data below 6.5l  \r\n  \r\n**Mileage \u0026 Year**  \r\n- Mileage has positive skewed distribution\r\n- Year has negative skewed distribution\r\n- We're gonna handling the outliers  \r\n  \r\n### Check linearity  \r\n\u003cimg src=\"./image/bar_car2.jpg\" alt=\"\" width = \"700\"/\u003e  \r\n  \r\n- OLS assumption of linear regression is linear, but from the plot, price is exponentially distributed\r\n- good transformation in that case is a log transformation  \r\n \r\n### Check Multicollienarity    \r\n\u003cimg src=\"./image/data_car2.jpg\" alt=\"\" width = \"150\"/\u003e\r\n\r\n- Year has the highest VIF, drop 'Year'\r\n- VIF 'Year' = 10.35 \u003e 10 ==\u003e unacceptable\r\n\r\n### Linear Regression Model\r\n\r\n\u003cimg src=\"./image/bar_car3.jpg\" alt=\"\" width = \"400\"\u003e\r\n\r\n- For high prices, we have a higher concentration of values around the 45 degree line ==\u003e Our model is very good at predicting higher prices.\r\n\r\n\u003cimg src=\"./image/accuracy_car.jpg\" alt=\"\" width = \"250\"\u003e\r\n\r\n- Our model is explaining 75% of the variablity of the data, relatively good result\r\n\r\n\u003cimg src=\"./image/data_car3.jpg\" alt=\"\" width = \"250\"\u003e  \r\n  \r\n**Weights Interpretation :**  \r\n- A positive weight shows that as a feature increases in value, so do log_price and 'Price' respectively\r\n- Example : EngineV, the bigger the Engine volume, the higher the price\r\n- A negative weight shows that as a feature increases in value, log_price and 'Price' decrease\r\n- Example : Mileage, the more a car is being driven, the lower the price gets\r\n- Dummies are compared with the benchamrk dummy\r\n- A positive weight shows that the respective category is more expensive than the benchmark\r\n- Example : respective category (BMV) is more expensive than the benchmark (Audi)\r\n- A negative weight shows that the respective category is less expensive than the benchmark\r\n- Example : respective category (Mitsubishi) is more expensive than the benchmark (Audi)\r\n- The bigger the weights the bigger the impact\r\n- Mileage is te most prominent feature in the regression. It is more than twice as important as Engine Volume\r\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnickenshidqia%2Fpredict_price_of_used_car_using_linear_regression","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnickenshidqia%2Fpredict_price_of_used_car_using_linear_regression","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnickenshidqia%2Fpredict_price_of_used_car_using_linear_regression/lists"}