Value-At-Risk (VAR) is a statistical technique that is used to enable the firm to measure and quantify the financial risk within a firm or within an investment portfolio over a specific time frame. Risk manager uses this technique to effectively control the level of risk undertaken by the firm. The level up to which a firm can easily absorb risk is ensured by the manager so that there may not be a possibility of expected losses.
Three general approaches are used to obtain effective portfolio loss distribution. The time frame, the amount of potential loss and the probability of that amount of loss are three variables that are considered by the model. Incremental VAR is used to measure likely worst case scenario for the portfolio under consideration as a whole within a given time frame (Benninga & Wiener, 33). In order to calculate the incremental VAR the investor is required to have an idea about the standard deviation of portfolio and its rate of return. In addition to this rate of return of asset in question and portfolio share is also needed to be incorporated.
To estimate VAR three approaches as follows are used:
Variance-covariance (VCV), This approach assumes that risk factor returns are (jointly) normally distributed and there is linear dependence in the change in portfolio value and the risk factor returns involved.
The historical simulation, This technique assumes that the distribution followed by asset returns in the future will be same as the distribution they followed in the past.
Monte Carlo simulation, This technique is used when future asset returns are expected to be randomly simulated