A robust VaR model under different time periods and weighting schemes (original) (raw)
Related papers
Evaluating the Predictive Performance of Value-at- Risk (VaR) Models on Nordic Market Indices.
With the course of financial markets becoming more global and complex, the need for effective risk management has become increasingly important for firms and financial institutions ever since. Value-at-risk (VaR) is one of the most widely accepted risk management tools to estimate market risks. This thesis applies 3 methods, namely Normal, Historical and Exponentially Weighted Moving Average (EWMA) to estimate VaR models for 3 major Nordic indices, namely OMXH25 (Finland), OMXS30 (Sweden) and OMXC20 (Denmark). The market risks are estimated both at 95% and 99% confidence levels. To evaluate the predictive performance of the VaR models, this thesis applies 8 backtesting methods that primarily conduct several frequency, independence and joint tests to evaluate the VaR models. This thesis evaluates the predictive performance of the VaR models, firstly, for the entire period of 2006-2013 which is considered as the benchmark year, and then, individually for 2008 and 2010 which are considered as crisis and tranquil year respectively. The empirical results show that the Nordic markets behave somewhat similarly when exposed to global market conditions. EWMA VaR model performs better and Normal VaR model performs worse at estimating market risks. The VaR models are mostly accepted at 95% confidence level by most backtesting methods. Besides, the VaR models perform poorly when exposed to extreme events such as global financial crisis in 2008. Otherwise, VaR models performs relatively accurate in normal market conditions.
Testing Applicability of Value at Risk Models in Stocks Markets
Mediterranean Journal of Social Sciences, 2014
This paper evaluates the forecasting performance of Value at Risk (VaR) method based on two wide spread approaches, historical simulation and Risk Metrics, before and after the sub-prime crisis in the context of developed and emerging capital markets. We present results on both VaR 1% and VaR 5% on a one-day horizon for Belex 15 and SAX. For comparative purposes, the paper also focuses on the DJIA and the STOXX Eastern Europe Total Market Index, an index representative of emerging European stock markets. In order to validate accuracy of VaR results we employ different back test techniques. Results indicate that the relative performance of VaR as a measure of market risk significantly underestimates the true level of market risk in Serbian stock market, in contrast to Slovak, where standard VaR approaches accurately capture market risk exposure. Results also provide evidence that the characteristic of stock markets and their asset returns in combination with the desired confidence level and risk horizon determine how well a certain approach performs on a certain stock market.
A Comparative Performance of Conventional Methods for Estimating Market Risk Using Value at Risk
This paper presents a comparative evaluation of the predictive performance of conventional univariate VaR models including unconditional normal distribution model, exponentially weighted moving average (EWMA/RiskMetrics), Historical Simulation, Filtered Historical Simulation, GARCH-normal and GARCH Students t models in terms of their forecasting accuracy. The paper empirically determines the extent to which the aforementioned methods are reliable in estimating one-day ahead Value at Risk (VaR). The analysis is based on daily closing prices of the USD/KES exchange rates over the period starting January 03, 2003 to December 31, 2016. In order to assess the performance of the models, the rolling window of approximately four years (n=1000 days) is used for backtesting purposes. The backtesting analysis covers the sub-period from November 2008 to December 2016, consequently including the most volatile periods of the Kenyan shilling and the historical all-time high in September 2015. The empirical results demonstrate that GJR-GARCH-t approach and Filtered Historical Simulation method with GARCH volatility specification perform competitively accurate in estimating VaR forecasts for both standard and more extreme quantiles thereby generally out-performing all the other models under consideration.
Back-testing the VaR risk measure: an empirical study
2017
This thesis verifies the worst case losses (Value-at-Risk) of financial returns over a specified time period with a certain level of confidence. The measurement of VaR hinges on the distribution of investment returns. In order to test whether or not the VaR model accurately represents reality, back-testing is carried out for one day horizon for a yearly rolling window. The standard VaR parametric model which is based on normal distribution of returns is tested on real data. Findings are that this model is better for historical VaR estimation for bigger exceedance probabilities such as 5%, 1%, 2% etc, while the Student’s t-distribution seems to be better for smaller exceedance probabilities such as 0.5%, 0.1% etc.
A Detailed Comparison of Value at Risk in International Stock Exchanges
2003
This work investigates the performance of different models of Value at Risk (VaR). We include a wider range of methods (Parametric, Historical simulation, Monte Carlo simulation, and Extreme value theory models) and several models to compute the conditional variance (exponential moving averages, GARCH and asymmetric GARCH models) under Normal and Student's t-distribution of returns. We analyse four European indexes (IBEX-35, CAC40, DAX and FTSE100), the American Dow Jones and S&P 500 indexes, the Japanese Nikkei 225 index and the Hong Kong Hang Seng index. We examine two periods: a stable period and a volatile one. To choose the best model, we employ a two-stage selection approach. First, we test the accuracy of different models of VaR. We use the unconditional and conditional coverage test, the Back-Testing criterion and the dynamic quantile test. A model survived if all tests indicated the model is accurate. With regard to the first stage, the best models are Parametric and Extreme value theory methods, when they use asymmetric and nonasymmetric GARCH models under Student's t-distribution of returns. Second, we evaluate the loss function of these models. We use several non-parametric tests to test the superiority of a VaR model in terms of the loss function. The result of the second stage indicates that the best model is a Parametric model with conditional variance estimated by asymmetric GARCH model under Student's t-distribution of returns. Nowadays the Parametric models are not as popular because some authors argue that the most conventional parametric specifications have failed in capturing some rare events. However, this paper shows that these models can obtain successful VaR measures if conditional variance is estimated with a GARCH model to capture the characteristic of the returns. This model is usually an exponential GARCH under Student's t-distribution of returns. JEL: G32, G11, C52.
Model-Based Stress Tests: Linking Stress Tests to VaR for Market Risk
SSRN Electronic Journal, 2000
Under the new capital accord stress tests are to be included in market risk regulatory capital calculations. This development necessitates a coherent and objective framework for stress testing portfolios exposed to market risk. Following recent criticism of stress testing methods our tests are conducted in the context of risk models, building on the VaR literature. First, to identify the most suitable risk models for stress testing, we apply an extensive back testing procedure that focuses on extreme market movements. We consider eight possible risk models including both conditional and unconditional models and four possible return distributions (normal, Student's t, empirical and normal mixture) applied to three heavily traded currency pairs using a sample of daily data spanning more than 20 years. Finding that risk models accommodating both volatility clustering and heavy tails are the most accurate predictors of extreme returns, we develop a corresponding model-based stress testing methodology. Our results are compared with traditional stress tests and we assess the implications for capital adequacy. On the basis of our results we conclude that the new recommendations for market risk regulatory capital calculation will have little impact on current levels of foreign exchange regulatory capital.
South African Journal of Economic and Management Sciences, 2011
Accurate modelling of volatility is important as it relates to the forecasting of Value-at-Risk (VaR). The RiskMetrics model to forecast volatility is the benchmark in the financial sector. In an important regulatory innovation, the Basel Committee has proposed the use of an internal method for modelling VaR instead of the strict use of the benchmark model. The aim of this paper is to evaluate the performance of RiskMetrics in comparison to other models of volatility forecasting, such as some family classes of the Generalised Auto Regressive Conditional Heteroscedasticity models, in forecasting the VaR in emerging markets. This paper makes use of the stock market index portfolio, the All-Share Index, as a case study to evaluate the market risk in emerging markets. The paper underlines the importance of asymmetric behaviour for VaR forecasting in emerging markets’ economies.
Back-Testing Approaches for Validating Var Models
International Journal of Engineering Science Technologies
Value at risk (VaR) is one of the important market risk measures. It measures the possible potential loss on given investment in terms of value, with certain probability for certain time horizon. In this paper, our aim is to discuss different back-testing approaches to validate VaR models, and also test it the real market data. We back tested VaR of Nifty 50 index obtained by Variance Co-variance method, Historical simulation method, Monte-Carlo simulation, and cubic polynomial regression method. We have used Total exceptions by binary back-testing over entire population. we have also used Basel Traffic Light Zone Test, Kupiec POF-test, Kupiec TUFF-test, and Haas’ Mixed-Kupiec test and analyzed the above methods.
VaR Analysis for the Shanghai Stock Market
ipcsit.com
In this paper we investigated the relevance of the skewed Student's t distribution innovation in capturing long-memory and asymmetry features in the volatility of Shanghai stock markets. We also examined the performance of in-sample and out-of-sample value-at-risk (VaR) analyses using the FIAPARCH model with the normal, Student's t, and skewed Student's t distribution innovations. We found that risk managers and portfolio investors can estimate VaR and optimal margin levels most accurately by using the skewed Student's t FIAPARCH VaR models of long and short trading positions in the Shanghai stock market.
Parametric Value-at-Risk analysis: Evidence from stock indices
The Quarterly Review of Economics and Finance, 2012
We evaluate the performance of several volatility models in estimating one-day-ahead Value-at-Risk (VaR) of seven stock market indices using a number of distributional assumptions. Because all returns series exhibit volatility clustering and long range memory, we examine GARCH-type models including fractionary integrated models under normal, Student-t and skewed Student-t distributions. Consistent with the idea that the accuracy of VaR estimates is sensitive to the adequacy of the volatility model used, we find that AR (1)-FIAPARCH (1,d,1) model, under a skewed Student-t distribution, outperforms all the models that we have considered including widely used ones such as GARCH (1,1) or HYGARCH (1,d,1). The superior performance of the skewed Student-t FIAPARCH model holds for all stock market indices, and for both long and short trading positions. Our findings can be explained by the fact that the skewed Student-t FIAPARCH model can jointly accounts for the salient features of financial time series: fat tails, asymmetry, volatility clustering and long memory. In the same vein, because it fails to account for most of these stylized facts, the RiskMetrics model provides the least accurate VaR estimation. Our results corroborate the calls for the use of more realistic assumptions in financial modeling.