Open AccessFORECASTING SHARE PRICES USING SOFT COMPUTING TECHNIQUES
Author(s) -
Vignesh Ganesan
Publication year - 2019
Publication title -
international journal of students research in technology and management
Language(s) - English
Resource type - Journals
ISSN - 2321-2543
DOI - 10.18510/ijsrtm.2019.722
Subject(s) - autoregressive integrated moving average , share price , python (programming language) , dividend , market share , computer science , econometrics , artificial neural network , sort , order (exchange) , time series , economics , artificial intelligence , stock exchange , machine learning , finance , database , operating system
Background: For a long time, there has been a trend of trading of shares. Brokerage firms and dealers buy/sell stocks for clients and companies. Their work is based on knowing how the share price of the company will react in the market. Market/ share price predictions are useful as the investor/broker can attempt to predict the output in order to maximize his dividends or minimize his losses.
Methodology: R and Python tools are used to sort, segregate and process the data, and techniques/algorithms such as Genetic Algorithm, ARIMA, Artificial Neural Networks, and Linear Regression are used to forecast results of data. Along with the model data, external factors affecting share prices also be taken into account.
Findings: For each of the applied algorithms, their results are compared and the difference in output with the real-time values has been observed and recorded.
Implications: Using data mining techniques, an attempt is made to estimate a prediction model to help forecast share prices.