Value at Risk (VaR) – study notes 1

Value at risk (VaR) is a technique used to estimate the probability of portfolio losses based on the statistical analysis of historical price trends and volatilities. It is commonly used  banks and security firms.

VaR contains three components:

1. a relatively high level of confidence (typically either 95% or 99%)
2. a time period (a day, a month or a year) and
3. an estimate of investment loss (expressed either in dollar or percentage terms)

A common statement used in VaR: What is the most I can - with a 95% or 99% level of confidence -  expect to lose in dollars over the next month?

There are three common methods of calculating VaR:

1. The historical method
2. The variance-covariance method
3. The Monte Carlo simulation

The following is an sample implementation of a single stock VaR (download it and run the macro in your local PC).

https://docs.google.com/leaf?id=0B3dIVDUl8UFRNzMyMGNkMjktNGYzYi00OGU1LWFkOWMtZWE5YTJjYjU3NzVh&hl=zh_TW

I would not go into the detail of these three methods, but if you want to know more I suggest you may read John C. Hull reading. Or there is some good online tutorial on finance in the following youtube channel:
http://www.youtube.com/user/bionicturtledotcom

Reference: http://www.mathfinance.cn/value-at-risk/