Predicting Edwards Lifesciences Stock Prices Using Machine Learning Model

Abstract

Predicting stock prices in the stock market is difficult work because the price can fluctuate up and down throughout the trading day, influenced by so many factors. However, the researchers can also use some advanced algorithms to capture the trajectory of the stock market as much as possible. This paper chooses the adjusted closing price as the stock price and aims to predict it for the next trading day based on opening prices, highest prices, lowest prices, volume, and date. Exploratory data analysis is conducted to explore hidden relationships in our stock data. Then, four machine-based learning models: linear regression, K-nearest, support vector machine, and random forest are applied to make predictions for our outcome: adjusted closing price. By comparing the root mean squared error, the model with the best performance is selected and treated as one of the candidates for future stock price prediction. Although these models are relatively fundamental and may not fully capture the complexities of stock prices, they provide a solid foundation for future work.