This project aims to predict the selling prices of used Volkswagen cars. The goal is to build and evaluate a predictive model using various data analysis techniques in R.

The project is part of a knowledge assessment within the framework of studies in data analysis/data science.

• Data/: Storage for raw data files (e.g., Excel, CSV). Processed versions are also saved here.
• Models/: Saved models, such as trained regression models in .rds format.
• Script/: All R scripts, organized by purpose:
  • 00_packages.R: Setup and installation of required packages
  • 01_data_cleaning.R: Data preparation and cleaning
  • 02_eda.R: Exploratory data analysis
  • 03_data_modeling.R: Creating the regression models
  • 04_model_diagnostics.R: Evaluating model performance
  • 05_model_selection.R: Comparing and selecting the best model
  • 06_model_testing.R: Testing the selected model
  • 07_predict_new_data.R: Using the model for predictions
  • SCB API: Use to download data
• Visualizations/: Finalized graphs and charts (PNG or PDF format).

# Core data manipulation and visualization library(tidyverse) library(ggplot2) library(readxl) library(stringr) # Data retrieval library(httr) library(jsonlite) library(pxweb) library(lubridate) # Analysis and modeling library(corrplot) library(caret) library(MASS) library(lmtest) library(car)

See Script/install_packages.R for installation of necessary packages.

1. Open Kunskapskontroll.Rproj in RStudio.
2. Run the scripts in Script/ in the following order:
   • 00_packages.R
   • 01_data_cleaning.R
   • 02_eda.R
   • 03_data_modeling.R
   • 04_model_diagnostics.R
   • 05_model_selection.R
   • 06_model_testing.R
   • 07_predict_new_data.R

For questions, please contact: [email protected]