نشریه تحقیقات مالی
پیاپی 50 (تابستان 97)

One of the main concerns of the market regulators is the prediction of the effects of these new strategies on the market due to the heterogeneity of the agents, rational boundary and behavioral factors in the investors’decision making. The Iranian stock market has always been fluctuating; therefore,awareness of the effects of strategies before they are implementedwill help regulators to market more effectively. The main objective of this research is to create an artificial market according to the Iranian stock market so that different scenarios can be simulated. One of these emerging areas, which emphasizes the impact of social sciences, cognitive sciences and behavioral sciences on operational research, is "Behavioral Operations Research" that helps us solve real-world problems.In this research, considering modeling based onagent-based capabilities, shareholders’ capabilities, bonds including different types of stocks and risk-free papers, and trading rules. In this artificial market in each trading period, in accordance with the trading strategy and learning procedures, the agents intendto buy and sell. Eventually they worked as the market makers, in accordance with the auction mechanism, and began to execute orders and perform clearing and settlement operations. In order to examine the validity of the model, the statistical output of this market wasadapted to the statistical characteristics of the financial markets and, after validating the model with the scenario, simulation of the research questions were done. In this research, the scenarios for eliminating the range of price fluctuations and elimination of the informed stakeholders and their effects on stock prices were reviewed. According to the simulated scenarios of the Iranian stock market, due to its immature nature, eliminating controlling mechanisms such as the range of price fluctuations, in the short term the Market willbe highly instable, but in the long run the market tends to be more efficient.

This contribution proposes an order placement strategy which can be run on simulating continuous financial markets, within an agent-based model framework. In order to improve the efficiency of price discovery, the order placement decision is given by an optimization model which minimizes the risk adjusted execution cost, taking into consideration relevant market microstructure factors such as market impact. The trading behavior of the agents has been extracted from intraday LOB data of Foulad Stock in Tehran Stock Exchange. The market has been simulated for 30 days and the results indicated that the optimized ordering strategy, in terms of the average purchase price of the share, the average waiting time for the transaction of each share and the average volume of the order traded, had better performance in comparison to other strategies examined. We can claim that taking into consideration both non-execution risk and execution cost could raise the performance in comparison to other strategies based on the aggressive level of the traders.

The purpose of this study is to develop a new approach to select effective variables in predicting financial distress using experts’ judgment and decision-making algorithms. Twenty nine financial ratios of financially distressed manufacturing companies according to Article 141 of the business Law were selected and the same number of healthy firms have been randomly selected from the companies which were listed in Tehran Stock Exchange between 1385 and 1395 using audited financial statements of one, two and three years before getting distressed. Then, using the statistical test and Dematel and Todim Fuzzy decision-making algorithms, the best financial ratios and their respective importance coefficients were selected and the prediction of financial distress was made using a support vector machine. Paired T-test results showed that accuracy difference of proposed model in predicting financial distress has been statistically significant in 5% level comparing to Altman Model and Logistic Regression Method for the years t-1, t-2, and t-3. The findings of the study showed that the proposed model has a significantly better performance in predicting distress than the Logistic regression method and Altman model in one, two and three years before financial distress.

chooseing the optimal portfolio is an integral issue in investment. It is challenging to estimate the tock return rate because of the vague nature of future events. Hence, it is recommended to add diverdity to cover the related risks. According to Markowitz’s modern portfolio theory, assets return has normal distribution. But empirical evidences show that assets return distribution is not normal and we would consider higher moment and Entropy as diversification indexes in this paper. In this study, a new approach for polynomial idealistic programming which is basedon a mean-variance-skewness-kurtosis-entropy model is proposed and two measures of Shannon entropy and Gini-Simpson entropy are used. The results indicated that the proposed approach is well-suited, especially for portfolio models with higher moments. The findings showed that using entropy as a diversification index cannot cause any significant decrease in optimized values for other goals. Using Shannon entropy and Gini-Simpson entropy models can lead to an increase in return and Shannon entropy model can yield more diversification compared to Gini-Simpson entropy model.

Margin in derivatives markets, such as futures markets, is used as a means of controlling the risk of future prospective obligations. On the other hand, margin is regarded as trading expenses and as one of the factors influencing futures market. Accordingly, in this paper, the effect of margin changes on futures contracts has been studied regarding Iran mercantile exchange in terms of return, risk and liquidity from 2016 to 2017. To do so, three approaches were used: estimating separate equations, seemingly unrelated equations and vector auto regression along with impulse response analysis. Based on the results, margin changes had a significant effect on returns and liquidity, but no significant effect on volatility. Also, the effects of margin change shocks on all three variables of return, volatility and liquidity were not stable, and more specifically the rate of damping of this effect on liquidity is less than the effect on the other variables. In other word, margin changes affect futures market return and liquidity, yet this is not a stable effect. These results can be helpful for futures market policy makers and investors.

The purpose of this study is to investigate the effect of financial distress on the relationship between earning volatility and capital structure decisions using Structural Equations Modeling (SEM) Approach. For this purpose, a sample of 82 companies were selected among the companies accepted by Tehran Stock Market between 2006 and 2017. To measure the moderating effect of financial distress, the sample companies were classified into two groups based on the KZ model. To measure the earning volatility, the researchers used observable variables such as coefficient of variation of ROE, coefficient of variation of OI divided by total assets and standard deviation of the percentage change in operating income. Also, we used three measures of total debt divided by total assets, total debt divided by book value of equity and total debt divided by market value of equity to measure the capital structure value. After ensuring the acceptable process of the research measurement and structural models, the results indicated that earning volatility has a significantly negative impact on capital structure decisions and financial distress significantly affect the relationship between earning volatility and capital structure. Moreover,such a relationship is proved stronger in the unconstrained companies group. The effect of earning volatility on capital structure decisions is stronger in sound Companiescompared to financially distressed holding Companies.

The aim of this study is to predict trend in stock using both analytical methods of stock prediction and intelligent machine learning methods on the case study of the Tehran Stock Exchange index. The proposed method consists of the following steps: at first, required data are collected. Afterwards, the data are evaluated using 25 analytical methods certified by Tehran stock exchange, Inc. Then, 10 highest rank methods are selected based on feature selection technique leading to a decrease in dimensions. The output of the final step is given to five intelligent machine learning methods, i.e., linear support vector machines, Gaussian kernel support vector machines, decision trees, Naïve Bayes and K nearest neighbors. Eventually, majority voting approach is used to make the final decision. The advantage of the proposed technique is the flexibility to use any technical analysis methods which means there is almost no limitation for this approach. Moreover, the feature selection technique is utilized for technical analysis and these methods are prioritized.