Predictive Analysis of AAPL Stock Trend by Random Forest and K-NN Classifier

Abstract

While perfection in stock market prediction is impossible to accomplish, minimizing the investment risk utilizing data mining classifier to forecast the Stock Market trend becomes one of the most popular fields to research. This paper analysis and forecast AAPL ’s daily stock trends by utilize two most frequently use data mining algorithm K-Nearest Neighbor and Random Forest Classifier. By comparing the accuracy of the two model to finds out which model would best estimate the AAPL’s stock price. Both Algorithm utilize simple rolling average as the predictors to increase the accuracy of the prediction. The result points out that random forest classifier has 61% of the total accuracy which is greater than K-Nearest Neighbor (KNN) with only 53% of accuracy in total for forecasting the AAPL stock trend. Random Forest Classifier have a better performance in predicting the AAPL’s stock market compared to KNN, however it may show a different result when it comes to other stock data.