Abstract

Stock price prediction is a challenging task because stock market is dynamic, unpredictable, noisy and volatile in nature. To address these challenges, various machine learning algorithms and models are developed to identify patterns in the stock price movements. Stock forecasting is so appealing for both institutional and individual investors. For institutional investors, an increase of just a few percentage points can increase profit by millions of dollars. Individual investors also seek profits through market participation by investing spare money in market. For a long time, researchers have worked on advanced intelligent techniques based on either technical or fundamental analysis of stocks to identify patterns in the stock price movements using advanced data mining techniques. Different models, including linear regression, Autoregressive Integrated Moving Average (ARIMA), Random Walk Theory (RWT), Moving Average Convergence/Divergence (MACD), Support Vector Machine (SVM), Autoregressive Moving Average (ARMA), Random Forest (RF), neural networks such as Artificial Neural Network (ANN), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN) and deep neural networks like Long Short Term Memory (LSTM) have been used for stock price prediction and shown promising results. In this paper, ARIMA model and SVM mode are applied to visualize, analyze and predict Apple Inc. closing stock price between Jan 01, 2012 and Jan 01, 17, 2023, I focus on two specific machine learning techniques, namely ARIMA mode and SVM model and applied them to visualize, analyze and predict Apple Inc. The results displayed that ARIMA model is better than SVM model with smaller MSE and high R-squared.