Analyse the Lack of Accuracy in Stock Price Prediction using Novel Linear Regression Compared with Support Vector Classifier to Improve Accuracy

The major objective of this essay seems to be to establish that Novel Linear Regression can forecast stock price more effectively than Support Vector Classifiers. With the results of a new analysis and a threshold of 0.05 percent, data has been gathered from several internet sites, a confidence interval of 95 percent, and mean and standard deviation. The classification is done by implementing a number of data points n = 10 in Novel LR and data had been collected n = 10 in SVC algorithms. In comparison to the SVC algorithm's accuracy of 45.73 percent, the analysis of the findings reveals that Novel Linear Regression has a high accuracy of 85.98 percent. The distinction between both the research groups economically unimportant (p = 0.921; p>0.05). It demonstrates because the Novel LR algorithm appears to produce better precision when predicting a customer's stock price than the Support Vector Classifier approach.