A Method To Find Historical VAR For Portfolio That Follows
S&P CNX Nifty Index By Estimating The Index Value
Ramesh. K.V.N.M* Email:firstname.lastname@example.org
Developer/Executive BCGS Mobile #: +65 83221677
60B Orchard Road
The Atrium@Orchard #10-00Singapore-238891
Abstract: Financial institutions face the important task of estimating the controlling their exposure to market risk, which arises through different risk factors in their portfolio. Measurement of market risk has focused on a metric called Value At Risk (VaR). VaR quantifies the maximal amount that may be lost in a portfolio for a given period of time,at certain confidence level. For large portfolios the risk factor can be taken as an index. In this paper we come up with a method of estimating Historical VaR for a portfolio that reflects the S&P CNX Nifty index at any point of time. We assume that the value of the index X (t) is independent of time and the distribution of X (t) is not necessarily gaussian.
Inimplementing firm-wide risk management there are two big challenges one is to implement interfaces to all the different front-office systems, back-office systems and databases, in order to get the portfolio positions and historical market data into a centralized risk management framework. The second challenge is to use the computed VaR numbers to _________________________________
* Working as aDeveloper/Executive (Commodities ) Barclays Capital Singapore
actually control risk and to build an atmosphere where all participants accept the risk management system.
The main contribution of this paper is to introduce a model, which predicts the boundaries of S&P CNX Nifty index,
this can be used for calculating the VaR of a portfolio, which follows that index.
The time series of an index is acomplex process. A piece of index sequence sampled over a time scale should be modeled as a non-stationary segment. This means a sequence of index values sampled over an arbitrary sampling period may be modeled as a non-stationary time sequence.
The rest of the paper is organized as follows, section 2 provides the Literature survey section 3 describes the Statistical Modeling for VaR describingthe model and approximation decisions for calculating VaR. Section 4 describes the characteristics of the Index, section 5 describes the DLF based prediction algorithm and VaR approximation, sections 6 & 7 explains a method to reduce the estimated VaR without forfeiting the confidence level by estimating the exact value of index in section 6 and using that in estimating VaR in section 7,Observations and results are discussed in sections 8 & 9 respectively. Section 10 gives the conclusions of the work done in this paper.
2 Literature Survey
Historical VaR is a better methodology to use if you cannot determine the distribution of your return series. As described in the papers given in references by “Simone Manganelli” and Robert F. Engle”, “Butler” and “Schachter” and by “EdgardoCayon Fallon” and “Julio Sarmiento” we rank all of your past historical returns in terms of lowest to highest and computing with a predetermined confidence rate what your lowest return historically has been. This means if you had 100 past returns and you wanted to know with 95% confidence what's the worst you can do, you would go to the 5th data point on your ranked series and know that 95% of thetime you will do no worse than this amount.
Historical VaR seems way too simplistic and in fact that is the biggest criticism of the methodology. Without a distribution to help determine future returns, you are assuming that the past will exactly replicate the future, which is very unlikely in itself. The strengths of the method are that all past data has been fully incorporated in the risk...
Lire le document complet
Veuillez vous inscrire pour avoir accès au document.