- Home
- Volume 85 - Année 2016
- Actes de colloques
- Special edition
- Prediction of Economic Value Added status of Tehran Stock Exchanges by using Genetic Algorithm
View(s): 1037 (37 ULiège)
Download(s): 397 (1 ULiège)
Prediction of Economic Value Added status of Tehran Stock Exchanges by using Genetic Algorithm
Attached document(s)
original pdf fileAbstract
The main purpose of this article is to present a model to predict the status of economic value added of Tehran Stock Exchanges by using Genetic Algorithms.
Investors’ concern of the return of principle as well as profit of their investment has led us to forecast status of economic value added as a basis to evaluate companies’ performance. Predicting the status of economic value added is one of the ways that can be used to exploit investment opportunities and also to avoid waste of resources. Firstly, necessary warnings can alert companies toward their future performance, moreover investors can recognize good and unfavorable investment opportunities.
In this research, independent variables are defined as: Equity premium Reserve, R & D Reserve, Precautionary Reserve, Legal Reserve, Special Reserve, Capital expenditures, Interest Bearing Liabilities, Finance Costs, Capital, Sum of Equity, Long Term Receivables, Nopat, Return of Capital, Investing, Accumulated Profit and Loss, Earning after interest and tax, Receivable Financing Facility and Economic Value Added is as a Dependent variable. The studied population is the accepted companies in Tehran Stock Exchange for the period of 1385 to 1391.
At first, by using average test among the variables, those variables that have a significant difference in the two groups of firms with positive EVA and negative EVA are specified. To do so, independent sample T Test from SPSS software is used.
Then, using the techniques of Genetic Algorithms, Which is one of the best evolutionary algorithms, variables with greatest ability to distinguish between positive EVA and negative EVA are specified then the predictive model is derived.
Derived model from the algorithm enabled us to classify the companies with negative economic value added into the correct categories of negative and positive economic value added, one year prior to the occurrence of negative EVA, with an accuracy of 83%.