This thesis examines penalized logistic regression methods for high-dimensional data. It provides an overview of logistic regression and its use in binary classification problems. The thesis discusses how applying logistic regression to high-dimensional data can be challenging due to overfitting. It then introduces penalized logistic regression as a technique to address this issue by adding a penalty term to the loss function. The thesis will review different penalized logistic regression methods like L1 and L2 regularization. It will also apply these methods to real-world datasets and compare their performance.