Company Sales Prediction and Stock Exhance Investor Profit Prediction based on Company Balance Sheet Data


Erboy M. O. , Bulut N., Shakeri S., Yuzuk S., AKTAŞ M. S.

4th International Conference on Computer Science and Engineering (UBMK), Samsun, Turkey, 11 - 15 September 2019, pp.201-205 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/ubmk.2019.8907132
  • City: Samsun
  • Country: Turkey
  • Page Numbers: pp.201-205
  • Keywords: Financial Time Series Based Forecast, Algorithms Cause-Effect Relation Learning, Sales Forecast, Regression Based Analysis, Fama French Three Factor Model

Abstract

Estimation of periodic sales of companies traded in BIST is very important for investors in transactions such as stock purchase. It is known that companies' balance sheet data is an important factor in these estimates. We see the lack of a comparative study of the success of the methods used to estimate the sales status of companies on balance sheet data. In addition, there is a need for methodologies that examine the returns of investors from the stocks of the companies they invest. For this purpose, the Fama French 3 Factor financial model, which reveals the investor's return as a result of stock trading transactions and the factors affecting it, has been studied within the scope of this research. Within the scope of this research, the accuracy of the algorithms for the estimation of company sales on BIST Balance Sheet data were tested and the factors affecting the returns of investors from stock trading transactions were examined. The tests show that the methods based on the cause-effect relationship produce the most accurate results in the estimation of company sales.