This paper studies four ARCH type models including ARCH, GARCH, EGARCH and TGARCH at Value at Risk (VaR) estimation. The four models were applied to daily Tehran stock market data to assess each model in estimating one day Value at Risk at various confidence intervals. Our findings suggest that for the daily return of Tehran Stock market, which exhibit fat-tailed and leptokurtic features, the VaR estimates generated by the GARCH-T models have good accuracy at high confidence levels.

Objective: Nowadays, the measurement of the risk of the marketplace has a significant effect on investments; however, the inadequate evaluation of this risk will cause a financial crisis and possible bankruptcy. One of the typical approaches to measure this risk is the probability-based risk measurement method, known as Valueat-Risk (VaR), for estimating and backtesting of which there are various methods. The purpose of this paper is to put forward a comprehensive test for backtesting and analyzing the sensitivity of VaR based on the number of samples (n) and confidence levels (N). Methods: First, the VaR of Tehran Stock Exchange data was estimated by applying GARCH-Copula, DCC, and EVT. Next, by using the multinomial backtesting in two steps the accuracy of VaR estimation and ranked the models were tested. Thereafter, considering the number of samples (n) and the confidence levels (N), the sensitive analysis of the backtesting result demonstrated the accuracy of the estimated VaR by selecting the most appropriate parameters. Results: Sensitive analysis findings indicated that in all three models, increasing the parameter "N" will result in an increase in the error rate. On the other hand, sensitive analysis of parameter "n" proved that its value depends on the technique used to estimate VaR, but generally, any increase in it leads to validation of VaR estimation models. The results also showed that according to the EVT method, at least 29% of the data is required to be used as a test sample in VaR estimation; however, the amount is equal to 22% in the DCC and GARCH-Copula methods. Conclusion: The result of the sensitivity analysis indicated that the reliability of different estimating VaR techniques relies on "n" and "N" parameters and different amounts of these two parameters can generate inaccurate and uncertain outcomes for each model. In addition, ranking these methods by using the loss function, GARCHCopula, EVT and DCC methods ranked first to third, respectively.

Due to the increasing growth of social networks in recent years, investors in various markets, in addition to reviewing and analyzing classic market information, also pay attention to news and information published on social networks. By examining and evaluating the relationship between news and information published on social networks and changes in stock prices, it is possible to understand the impact of the information on stock prices and predict the future trend. In this article, using the methods of sentiment analysis and text mining, the impact of public thoughts and feelings caused by news on the Internet and cyberspace on stock prices is examined. The information used in this research includes content published on the social network Twitter about stocks and real stock price data of the top 5 companies on the US stock exchange. Using the presented method, general feelings about a text are estimated and a general score is considered for it. Then, using back testing methods and adopting different trading strategies, the impact of these emotions on the share price trend will be examined and the obtained results will be compared with and without the effect of emotion analysis. According to the results of this study, the effectiveness of strategies based on sentiments analysis is significantly higher than technical analysis-based methods.

Financial market developments make it more important to measure market risks correctly. In this paper we investigatethe forecasting accuracy of different historical simulation models in relation to the risk measure expected shortfall and in comparison to established parametric models.we used historical simulation, mirrored historical simulatin, volatility weighted historical simulation, filtered historical simulation and GARCH (1, 1) models. The data that we used consists of Tehran stock exchange market index from 2010 to 2014.Christofferson backtest used for value at risk and mc neil & frey backtest used for expected shortfall. According to unconditional coverage backtesting, mirrored historical simulation model was rejected and others were accepted and according to independence backtesting all models were accepted thus the christoferson backtest will omit the mirrored historical simulation model and According to mc neil and frey backtest all models were accepted and finally the model confidence set procedure showed that semi parametric models are best models to forecast expected shortfall.

The expansion of the capital market and the reduction of interest rates on commercial banks has made that investing in dominate shares as one of the most important opportunities for obtain gain on investment, which requires risk acceptance. In this paper, the goal is to extract the residual values of the logarithmic return of the Tehran stock exchange index using the CIPRA model. Then, using the extreme value theory, the extreme value model was obtained for these residual. Extreme value theory is a good approach to the estimation of high and low tails and measure such as Value Risk (VaR). In order to determine the performance of this method, another model was compared with this model using the two indexes include Tehran Stock Exchange and the Top 50 Industry Index at 99 and 99. 5% for the estimated value of risk. The results of the backtesting show that the EVT-CIPRA approach works better.

Value at Risk (VaR) plays a central role in risk management. There are several approaches for the estimation of VaR, such as historical simulation, the variance-covariance and the Monte Carlo approaches. This work presents portfolio VaR using an approach combining Copula functions, Extreme Value Theory (EVT) and GARCH-GJR models. We investigate the interactions between Tehran Stock Exchange Price Index (TEPIX) and Composite NASDAQ Index. We first use an asymmetric GARCH model and an EVT method to model the marginal distributions of each log returns series and then use Copula functions (Gaussian, Student’ s t, Clayton, Gumbel and Frank) to link the marginal distributions together into a multivariate distribution. The portfolio VaR is then estimated. To check the goodness of fit of the approach, Backtesting methods are used. The empirical results show that, compared with traditional methods, the copula model captures the value more successfully.

Value at Risk (VaR) plays a central role in risk management. There are several approaches for the estimation of VaR, such as historical simulation, the variance-covariance and the Monte Carlo approaches. This work presents portfolio VaR using an approach combining Copula functions, Extreme Value Theory (EVT) and GARCH-GJR models. We investigate the interactions between Tehran Stock Exchange Price Index (TEPIX) and Composite NASDAQ Index. We first use an asymmetric GARCH model and an EVT method to model the marginal distributions of each log returns series and then use Copula functions (Gaussian, Student’ s t, Clayton, Gumbel and Frank) to link the marginal distributions together into a multivariate distribution. The portfolio VaR is then estimated. To check the goodness of fit of the approach, Backtesting methods are used. The empirical results show that, compared with traditional methods, the copula model captures the value more successfully.

Value at Risk (VaR) is the maximum loss which could be incurred within a given time horizon, except for a small percentage, that its application has sharply increased after the 90s. Parallel to the increase in usage of value-at-risk in risk management areas, validation of VaR measures has gain great importance. In prevalent back testing approaches, returns which are yielded from VaR estimators are not regarded as a criterion. It's may not be desirable for the investors who emphasize on return more than the risk. What distinguishes this study from other researches in the field of back testing VaR estimation models is the simultaneous consideration of actual return and loss(CVaR) which were yielded from VaR estimators as criteria of risk and return that are the primary basis for financial studies. On the other hand, due to relativeness of risk and return in terms of investors, we considered the weight of these two indexes as fuzzy. In this paper, we constitute and optimize our risky portfolio with safety-first investor's rule. We need to estimate quantile of risky portfolio's return in objective function of safety-first investor's rule to optimize the portfolio. VaR estimators were used to calculate it. On the other hand, given the non- convexity of VaR function and also other reasons, we applied one of the most popular meta-heuristic models namely genetic algorithms for optimization. Our findings show that GEV and HS models are more conservative than parametric models (t-student and normal) and also have better performance in portfolio optimization. The empirical findings also indicate that safety-first investor will choose significantly different amounts of borrowing. Thus, the scale of the risky portfolio and the amount borrowed is diverse across methods. There is another interesting finding. Despite the computational simplicity of historical simulation method, it has shown the best performance of all.

The recent global financial crisis causes that financial markets participants provide an acceptable framework for their risk coverage. One of the most important risk measures for this purpose is value at risk (VaR) which is intended in finance literature in the past two decades. In general there are three approaches including parametric, nonparametric and semi-parametric techniques is used for estimating of VaR. This paper presents a new method that is named Markov Chain Monte Carlo (MCMC) simulation which is based on reproduction and generation of data such as Monte Carlo simulation methods. But, in this new method, data production is done in basis of Metropolis-Hastings algorithm. Considering quantile of generated returns distribution, VaR is calculated. Next, one day ahead value at risk of Tehran Stock Exchange indices for 200 future days are forecasted via this new method and also the accuracy of estimated VaR is evaluated by conditional and unconditional coverage backtesting statistics. Empirical results of this paper indicate that MCMC method for estimation and forecasting VaR of TSE indices has a reliable performance.