Stock Value Estimation using Linear Regression
Rama Chandra Naradasu1, Dharma Sri Harsha Tontepu2, Subramanyam Kodukula3
1Rama Chandra Naradasu, CSE department, Koneru Lakshmiah Educational Foundation, Vaddeswaram, India.
2Dharma Sri Harsha Tontepu, CSE department, Koneru Lakshmiah Educational Foundation, Vaddeswaram, India.
3Dr.Subramanyam Kodukula, CSE department, Koneru Lakshmiah Educational Foundation, Vaddeswaram, India.
Manuscript received on April 30, 2020. | Revised Manuscript received on May 06, 2020. | Manuscript published on May 30, 2020. | PP: 1443-1446 | Volume-9 Issue-1, May 2020. | Retrieval Number: F9853038620/2020©BEIESP | DOI: 10.35940/ijrte.F9853.059120
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Estimating the stock value is a tough task, because it depends more on value of stock and there`s no exact variable which is able to guess exactly a value of a stock every other day. What so Ever, Efficient Market Hypothesis said a stock vaue is crucially dependent upon new information. Many sources of info is the choice of the person in the social media.The choice of public on factory outlet from a specific firms can determine the stability of that particular firm and thus affect the decision of the many members to buy the company’s stock. When using opinion as an important data, an appropriate analysis of that opinion is necessary. One of the well know example of using opinion as an important data is anbrief note of sentiments.Analysis of sentiment is a way to determine emotion within the choice of public about some reason, in the given case some of thecorporations goods. There is some way of analyzes of the sentiment required to guess stock prices. Bollen concludes on his research that with 87.6 per cent accuracy, interestin social networking site such as Twitter can guess DJIA interest. This shows a clear relationship between the analysis of sentiments and the stock values. Our goal in this research is to use simple sentiment analysis to forecast Indonesian stock market. Naïve Bayes and the algorithm Random Forest are used to identify tweet to measure a company’s opinion. Sentiment analysis are used to display the stock value for the product. The prediction model is built using linear regression approach. Our research shows that predictive models are using before stock prices and hybrid feature as predictor provide the accurate prediction with determination coefficient of 0.9989 and 0.9983.
Keywords: Sentiment Analysis, Linear Regression, Stock Price, Supervised Learning, Efficient Market Hypothesis, Random Forest, Prediction.
Scope of the Article: Machine Learning