A Prototype Classification Algorithm for Stock Price Prediction using Optimized Variance of Attributes

Authors

  • Jayshree Bhangre, Anurag Jain

DOI:

https://doi.org/10.17762/msea.v71i4.672

Abstract

The advancement of machine learning algorithms increases the prediction of stock prices. The enhanced prediction ratio moves the stock market into the mode of stability and attacks investors to invest in the stock market. This paper proposed a prototype classification algorithm for the prediction of stock price. The prototype classification algorithm bags a clustering algorithm and classification algorithm. The bagging of the algorithm improves the data normalization factors and reduces the variance of attributes of stock data. The employed support vector machine uses rank factors of an improved clustering algorithm and increases class voting. The proposed algorithm is implemented in MATLAB tools for data analysis. The results of the proposed algorithm compared with SOM neural network model and SVM. The study of effects represents that the proposed algorithm is very efficient in SBI bank data sets. The evaluation of performance estimates in RMSE, NMSE, MAR and MI. 

Downloads

Published

2022-09-01

How to Cite

Jayshree Bhangre, Anurag Jain. (2022). A Prototype Classification Algorithm for Stock Price Prediction using Optimized Variance of Attributes. Mathematical Statistician and Engineering Applications, 71(4), 1574–1586. https://doi.org/10.17762/msea.v71i4.672

Issue

Section

Articles