Mohammad Hossein Setayesh; Mostafa Kazemnezhad
Abstract
The Purpose of this research is investigating the usefulness of variables (dimension) reduction methods (selection and extraction) in stock returns of the companies listed on Tehran Stock Exchange (TSE). In this regard, through reviewing literature, 52 predictive features (variables) were specified as ...
Read More
The Purpose of this research is investigating the usefulness of variables (dimension) reduction methods (selection and extraction) in stock returns of the companies listed on Tehran Stock Exchange (TSE). In this regard, through reviewing literature, 52 predictive features (variables) were specified as the initial features based on the popularity in the literature and the availability of the necessary data. By using variables selection (relief) and variables extraction (factor analysis) methods, optimal variables (factors) are selected or extracted from initial variables. Subsequently, the stock returns of 101 firms listed on TSE from 2004 to 2013 were predicted utilizing decision tree and linear regression. The experimental results confirmed the usefulness of variables (dimension) reduction methods in stock return prediction and better performance of relief (relative to factor analysis). Furthermore, the results indicated that decision tree outperforms the linear regression.
M-B Bagherpour; M Bagheri; H Khadem; R Hosieni Pour
Volume 9, Issue 34 , July 2012, , Pages 103-128
Abstract
Tax revenue is one of the most important financial sources of the Government, which has a leading role in the economic development of each country. However, some companies try to avoid paying their correct taxes, which creates a problem called "tax evasion". The difference between the tax stated by the ...
Read More
Tax revenue is one of the most important financial sources of the Government, which has a leading role in the economic development of each country. However, some companies try to avoid paying their correct taxes, which creates a problem called "tax evasion". The difference between the tax stated by the company and the tax payment identified by the tax authorities (Tax Organization) can be considered as an example of tax evasion. At this time, Tax organization is applying traditional methods to deal with this challenge, which can reduce the Government's revenue and increase its expenses in the long-term. The objective of this research is to apply data mining techniques to examine the effects of financial and non-financial variables on tax evasion by the companies operating in automotive and parts manufacturing industry. The findings show that "assets to net revenue ratio", absolute value of interest expense to net revenue ratio", and board independence" increase and "net income (loss) to assets ratio", and company's performance (net income or loss) decrease the likelihood of tax evasion. These findings can help tax authorities in both policy making as well as conducting tax audit
