This repository contains an implementation of Ordinary Least Squares (OLS) regression in C++ along with a data loader to read CSV files and a utility to save the regression results to a CSV file.

The project includes three main components:

1. DataLoader: Reads a CSV file, ignoring the header, and separates the features and target variable.
2. OLS: Performs OLS regression, predicts new data, and saves the results to a CSV file.
3. Main Program: Demonstrates the usage of DataLoader and OLS classes.

• DataLoader.h and DataLoader.cpp: Implementation of the DataLoader class.
• OLS.h and OLS.cpp: Implementation of the OLS class.
• main.cpp: Example usage of the DataLoader and OLS classes.
• data.csv: Example CSV data file (you should provide your own data file).
• results.csv: Output file containing the regression coefficients and predictions.

• A C++ compiler that supports C++11 or later.
• CMake (optional, but recommended for building the project).

1. Using g++ directly

   g++ -o ols main.cpp DataLoader.cpp OLS.cpp

Ensure that you have a data.csv file in the same directory as the executable. The data.csv file should have the following format:

feature1,feature2,...,featureN,target
value1,value2,...,valueN,target_value
...

feature1,feature2,target
1,1,6
1,2,8
2,2,9
2,3,11

To use your own data, replace the data.csv file with your dataset. Ensure that the target variable is in the last column and that the file has a header row.

.
├── DataLoader.h
├── DataLoader.cpp
├── OLS.h
├── OLS.cpp
├── main.cpp
├── data.csv        # Example input file
├── results.csv     # Output file containing results
└── README.md       # This file

This project is licensed under the MIT License. See the LICENSE file for details.

Contributions are welcome! Feel free to open issues or submit pull requests.

For any questions or feedback, please open an issue or contact me via GitHub.