Analyzing algorithm precision for Stock Market datasets

Abstract

The analysis proposed in this paper is a result of observation from our extensive and rigorous study, for analyzing the algorithmic precision of supervised machine learning algorithms. The usage of machine learning algorithms is done in order to find the predictive capabilities for a handful models for time series analysis using stock market datasets. This is done by the usage of scientific tools available in Python such as Web API’s for Data retrieval, Pandas, Training and Testing methods. Here, stock market datasets are used for checking the accuracy of predictive precision by means of numerical data. The main aim is to identify which algorithm best suits for numeric analysis, in terms of accuracy and less erroneous results while prediction.