Value at Risk
VDMV Lakshmi
An attempt to encapsulate an estimate of the price risk possessed by a portfolio of derivatives
and other financial assets is Value at Risk (VaR). It provides the dollar amount by which the
value of a portfolio might change with a stated probability during a stated time horizon; the
time horizon might be- one day, one week or longer.
In other words,
Value at Risk: The maximum losses that an investment may incur given a confidence
level during a given period.
Ex: You have invested Rs.1,00,000. One day 1% VaR is 1000
Max loss that may occur is Rs.1000 at 99% CL (normal market conditions) during a day.
You may incur more than Rs. 1000 loss in remaining 1% (black swan or extra ordinary or
abnormal or unpredictable cases)
VaR Computation Methods:
1. Historical Method
2. Variance-Covariance Method
3. Montecarlo Simulation Method
Confidence level (Significance level) No. of SDs
99% (1%) 2.327 or 2.326 or 2.33
95% (5%) 1.645 or 1.65
1. The following are the returns from a stock for the last 260 days as per historical
observation:
S No Returns No of Days S No Returns No of Days
1 -5.0% 3 12 1.00% 20
2 -4.8% 2 13 1.40% 30
3 -4.0% 3 14 2.10% 20
4 -3.5% 5 15 3.20% 10
5 -3.0% 5 16 3.80% 34
6 -2.8% 2 17 4.50% 19
7 -2.0% 2 18 5.20% 43
8 1.5% 1 19 5.80% 23
9 -1.0% 3 20 6.40% 5
10 0.5% 10 21 7.10% 6
11 0.0% 5 22 7.40% 9
1
If your investment is Rs 1,00,000, compute 5% VaR for one day period using
historical method.
Sol. Historical Method:
Returns Prob Cum Prob Returns Prob Cum Prob
-5.00% 1.15% 1.15% 1.00% 7.70% 23.50%
-4.80% 0.77% 1.92% 1.40% 11.50% 35.00%
-4.00% 1.15% 3.08% 2.10% 7.70% 42.70%
-3.50% 1.92% 5.00% 3.20% 3.80% 46.50%
-3.00% 1.92% 6.92% 3.80% 13.10% 59.60%
-2.80% 0.77% 7.69% 4.50% 7.30% 66.90%
-2.00% 0.77% 8.46% 5.20% 16.50% 83.50%
1.50% 0.38% 8.85% 5.80% 8.80% 92.30%
-1.00% 1.15% 10.00% 6.40% 1.90% 94.20%
0.50% 3.85% 13.80% 7.10% 2.30% 96.50%
0.00% 1.92% 15.80% 7.40% 3.50% 100.00%
VaR as per historical method at 5% level = -3.5% ×100000= - Rs 3,500
2. Consider a position consisting of a Rs.400,000 investment in stocks and a Rs.500,000
investment in bonds. Assume that the daily volatilities of assets are 2% each and that
the coefficient of correlation between their returns is 0.35. What is the 8-day 1% VaR
of the individual investments and portfolio? Also observe if there is any benefit of
diversification.
Sol. 8-day 1% VaR of investment in stock = 8000 ×√ ×2.326 = Rs. 52,631
8-day 1% VaR of investment in bond = 10000 ×√ ×2.326 = Rs. 65,619
The standard deviation of the daily change in the investment in each asset is $8,000
and $10,000. The variance of the portfolio’s daily change is
Daily standard deviation of portfolio is =
√
√
=14832
8-day 1% VaR of investment in portfolio
= 14832 ×√ ×2.326 = Rs. 97578
Benefit of diversification = (VaR of Stocks + VaR of bonds) – VaR of portfolio
= Rs. 52,631 + Rs. 65,619 - Rs. 97,578 = Rs. 20,672
3. You invested in a portfolio with Rs.80,00,000 of stock A and Rs.60,00,000 of stock
B. The expected returns on the two shares have a bivariate normal distribution with a
correlation of 0.1. 1-day standard deviation of stocks of A and B are 30,000, and
60,000 respectively. What is the 3-day 99% VaR of the portfolio? What is the 3-day
1% VaR of the individual stocks and portfolio? Also observe if there is any benefit of
diversification.
Sol. 3-day 1% VaR of investment in stock A = 30000 ×√ ×2.326 = 120862
2
3-day 1% VaR of investment in stock B = 60000 ×√ ×2.326 =241725
One day standard deviation of portfolio is =
√
= 69713
3-day 1% VaR of Portfolio = 69713 ×√ ×2.326 = 280859
Benefit of diversification = (120862+241725) – 280859 = 81728
4. A US based company has long position of French Francs 500 mn on the spot market.
The volatility is 10% annualized and the exchange rate is 5. Compute one day value at
Risk (VaR). (Assume 250 working days in a year)
Sol. A US based company has long position of French Francs 500 mn on the spot market.
The volatility is 10% annualized and the exchange rate is 5. Compute Value at Risk
(VaR). (Assume 250 working days in a year)
Daily Volatility = 10/√ = 0.6325%
1 day 1% VaR = 2.326 ×0.6325% = 1.471195 ≈1.47%
1 day 1% VaR = 2.326 σ = 1.47% ×$ 100 mn = $1.47mn
5. An investment company has a portfolio of shares of Reliance Industries recently
purchased at Rs 800. Assuming the volatility is 20% p.a. Calculate one day VaR at
95% level of confidence.
Sol. An investment company has a portfolio of shares of Reliance Industries recently
purchased at Rs 500. Assuming the volatility is 20% p.a. calculate VaR at 95% level
of confidence.
Daily volatility = 20/250^0.5 = 1.265%
One Day 5% VaR =1.265%×1.645 = 2.0809%
One Day 5% VaR 2.0809%×800 = 16.698
Value of asset can be = 800 + 16.698 = 816.698 or = 800 - 16.698 = 783.302
