FINANCIAL ENGINEERING AND SECURITIES MANAGEMENT (PORTFOLIO MANAGEMENT)
Year:2013 | Volume:3 | Issue:13|Papers Count:8

Credit risk is one of the main sources of risk for credit assets, so that the banks and bond investors have been very careful about measuring accurate credit risk. The importance of this issue and the lack of empirical studies conducted in this area in the country is the main spring of doing this research.Using dynamic model of Merton and by gathering the data of four listed companies in metal sector from 1380 to 1390 dynamic Merton and Z score model was applied to determine the default probability of companies as well as testing the relation between default and free float rate of a company. Despite of many other typical models in this area, those stocks that do not trade in the market was also considered. The hypothesis of normal distribution of asset value and fitted the best possible distribution to data in Merton's model was questioned in the model. The results indicate that the higher free float rate can be considered as one of the successful factors for companies in long time.

The purpose of this research is to Forecast Free Cash Flow (FCF), as the measure of financial performance, by using the lowest possible information. The statistical population of this research consists of all the firms listed in Tehran Stock Exchange. By considering some conditions, 64 companies were investigated for the years 1999-2009. For predicting FCF Simple exponential smoothing model and Grey model is applied. Then based on the Mean Square Errors (MSE), the accuracy of the models is compared. The results demonstrate that the accuracy of the Grey model is more than the Simple exponential smoothing model, and at the confidence level of 95%, the difference between the mean of the MSE in the Simple exponential smoothing model and Grey model, is not significant, while at the confidence level of 90%, the difference between the mean of the MSE in two forecasting models, is significant.

The fluctuations of housing price in some countries including Iran during 2 recent decades is one of the fundamental challenges in housing market and Iran economy so that in a special period the considerable price rise in housing market would be created and in the other period reduction or constancy would be dominated on housing price highly. The fact that the price rise is rooted in economy fundamental conditions or is due to bulb can be led to various policies or decisions. Bulbs have negative affects on economy. Paying attention to the price bulb procedure in housing market is important because it lead to wrong resource assignment and would rise the paper activities level. The bulb decline that is followed by economy bulb can waste the high amount of wealth. In this research the bulb in Tehran housing market using of ARDL technique using of haphazard function would be explored. In this research the seasonal data from 1375 to 1389 would be employed. The results from the model show that the bulb is a short term subject and in the long term it is a fundamental component of housing price. The earth price and housing price in previous period is the major factor for determination of housing price bulb in Tehran.

In order to study the prediction of corporate bankruptcy using fuzzy neural network models have the ability to stock in a dynamic and uncertain environment, it makes possible.The fuzzy variable using different words to define each indicator has been identified And with each membership functions using neural networks to create a learning system has been attempted. Among the various models of neural networks, multi-layer perceptron network, the nipple, with the error back propagation learning rule is selected. Here are four fuzzy neural network is used.Bankruptcy in order to identify the parameters affecting the final variables in three major categories as inputs to the neural network and fuzzy liquidity, leverage and the market is considered.Neural networks and fuzzy outputs liquidity, leverage and market as the input fuzzy neural network to predict bankruptcy of companies can be entered and exhaust the stock of pharmaceutical companies shows that the probability of bankruptcy.Performance of neural networks in fuzzy environment, a fuzzy inference system Sugeno and J. Bell membership functions is evaluated. The research results obtained with a model to predict the optimal minimum error is presented.

One of the important issues in the capital markets and should be considered is portfolio selection. In this regard, the investors are being studied in order to select the best portfolio with respect to risk level and the return. Nowadays, for portfolio selection investors use lots of risk measures so that these risk measures have applied with respect to KNOWLEDGE level of investors and their ability to analysis financial data. Therefore, in this study that has been conducted in the context of Iran's capital market, we provided a single period multi objective mathematical model with probable constraint to measure portfolio's risk which combines the return measure with two risk measures (semi variance and absolute deviation). Hence, investors can accurately measure their desired portfolio's risk, considering the limitations associated with trading costs. In this way, they can achieve a portfolio with the highest return and lowest risk.

Return-Volume dependence have been of a great important in the financial economy field. But the stock trading volume is not usually considered when the time of analyzing the stock specifications by investors, or when they tend to distinguish a price for a stock.In this article, a copula approach is applied to determine the relationship between volume and return in Iran emerging equity market. Copulas are functions that join multivariate marginal distribution functions to form joint distribution functions. Final results show a specified, important, and asymmetric dependency between volume and return, in Iran equity market. In other words extremely high returns (large gains) tend to have the some pattern as extremely high volume has. But the extremely low returns (big losses) don't tend to be the some as extremely low volumes.

Generally, "The biggest risk in the capital market or ina portfolio (capital market, Bank ...) occurswhena sudden large change occur towards its unfavorable basket. It's essential for financial risk management knowing the probability that such case sareveryrare and estimated its consequences. These values (Extreme Movement) are located at the tail of the distribution function, and therefore they named "Extreme Values".In this study, we followed the distribution of Iran stock Exchange returns (TEDPIX and Industrial Index) in two different time periods. We are testing afat tail in two different time periods. Generalized Extreme Value Theory (GEV) results show there are fat tails in the distribution function of return for both indices and for both periods. Finally, the Back testing results for the VaR calculated with this approach show that the model for 100-daytime horizonhas better performance than the 50- daytime horizon. We use Statistics Lopez tocomparethe performance of these Approach models (GEV); with VaR calculated with model Risk metric models with assuming normal distribution for different confidence levels. We reached to this conclusion that the GEV has better performance, because focuses on tail distribution function more than others approaches.

Value at Risk (VaR) risk management concept is a very useful technique for analyzers to manage the risk. But considering that, this model cannot estimate the complete risk of the tails, so there are a lot practices to develop better models to cover this shortage. In this study we try to review the shortages of this model and to amend and modify it by using two concepts "Cornish Fisher expansion" and "Elliptical t-student distribution" to have a better estimation of the mentioned risk. In this study we introduce two new models that are solved by SA algorithm. The efficiency of such models is shown by efficient frontiers. The findings show considerable improvement of using these two models