Scribd Logo
Upload

Welcome to Scribd!

  • 7 views

    Entrade Historical Simulation VaR Methodologies

    Uploaded by

    shri.suyash

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Entrade Historical Simulation VaR Methodologies

    Uploaded by

    shri.suyash
    7 views4 pages

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    7 views4 pages

    Entrade Historical Simulation VaR Methodologies

    Uploaded by

    shri.suyash

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 4

    Overview

    Historical Simulation VaR Method utilizes historical time series of market prices to obtain a distribution of historical
    returns. It evaluates current portfolio(s) across a set of historical price changes, to yield a distribution of changes. In
    portfolio MtM value, it can obtain the value at a percentile (for example: VaR), under the assumption that the
    historical price changes is a good forecast (for the range of changes the portfolio could have in the future). This
    approach computes a VaR at the chosen confidence level. As a consequence, the correlations between the
    individual risk factors are already taken into account.

    Two main steps should be taken for calculating historical VaR:

    1. The first step valuates all instruments using MtM methodologies to represent the current known value of the
    portfolio.
    2. The second step estimates the future potential range of the MtM value at the future VaR horizon. This
    represents the distribution of unknown future value of the portfolio (as illustrated in Figure 1).

    The basic steps in Entrade Historical Simulation VaR calculation are as below:

    1. Set VaR calculation parameters including: number of historical days, time horizon, confidence level, and
    analysis date. A sample decaying factor can also be defined to use sample decaying feature.

    2. Select Portfolio(s) to be analyzed.


    3. Obtain all underlying trade position and risk factors in the portfolio(s).
    4. Compute current MtM value of all trades on the analysis date.
    5. Compute day-by-day price changes for each historical day starting from the analysis date as:
    6. Calculate simulated prices for each iteration as:

    where S0 is the price at the analysis day and Si is ith simulated price at time horizon.

    7. Revaluate the portfolio at time horizon for each iteration using the simulated prices.

    8. Calculate the portfolio profit and loss (P/L) for the iteration (i.e. the difference between the current market
    value of the portfolio and the simulated portfolio value at time horizon).

    9. If the sample decaying factor is not set to 0> <1, i.e. not using sample decaying, a percentile number N
    of confidence is calculated as below:

    N= +1

    Next, sort the PnLs calculated in step 8 in ascending order to give simulated P/L distribution for the portfolio as
    Figure 3:

    Then VaR is the PnL at the percentile N.

    10. If the sample decaying lamba is selected, the BRW (Boudoukh, Richardson, Whitelaw) sample decaying
    method will be used to exponentially decaying samples. This approach assigns exponential weight to each
    historical sample. It combines the strength of EWMA (Exponentially Weighted Moving Average) and historical
    simulation.

    The sample weight for each historical day is calculated as below:

    Where wi is ith sample weight from the analysis date and is the decaying factor. Figure 4 below illustrated

    weighting factors of 81 days with = 0.96.


    Next, sort the PnLs calculated in step 8 in ascending order to give simulated P/L distribution for the portfolio as
    Figure 3. Then the normalized cumulative weights of the sorted PnLs will be calculated.

    Then, VaR is calculated using linear interpolation from the 2 PnLs which Normalized Cumulative Weights are right
    above and under the confidence level.

    11. Draw PnL Histogram to illustrate PnL distribution as Figure 5.


    ® 2021 Enuit LLC. All rights reserved.
    Send feedback on this topic to Enuit LLC.

    You might also like