Netflix Stock Price Prediction: A Comparison of Linear Regression, k-Nearest Neighbor, and Decision Tree Methods

Abstract

The implementation of stock prediction has become a major popular topic. However, the effectiveness of different machine learning methods for stock prediction tends to vary. Linear regression, k-Nearest neighbors, and decision trees are three basic machine learning algorithms and are frequently used in the practice of stock prediction. This paper proposes these three methods and compares the effectiveness of applying them to predict the stock price of Netflix by evaluating the mean absolute error (MAE), mean squared error (MSE), and coefficient of determination (R-squared). The results of these three methods all display promising predictions, but the linear regression model tends to outperform in this Netflix prediction attempt. This prediction practice finds patterns and trends in past data so as to help investors make more informed decisions in the volatile financial market.