Scribd
Upload
4 views17 pages

Value at Risk

Value at Risk (VaR) quantifies the minimum expected loss over a specified time period under certain market conditions, with a 5% Monthly VaR indicating a 2.8% loss or Rs. 2800. Various methods for estimating VaR include historical simulation, parametric method, and Monte Carlo simulation, each utilizing different data and assumptions. Back-testing is essential to validate VaR calculations, particularly to identify if actual returns deviate from the expected normal distribution.

Uploaded by

sairam16
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
4 views17 pages

Value at Risk

Value at Risk (VaR) quantifies the minimum expected loss over a specified time period under certain market conditions, with a 5% Monthly VaR indicating a 2.8% loss or Rs. 2800. Various methods for estimating VaR include historical simulation, parametric method, and Monte Carlo simulation, each utilizing different data and assumptions. Back-testing is essential to validate VaR calculations, particularly to identify if actual returns deviate from the expected normal distribution.

Uploaded by

sairam16
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Value at Risk (VaR)

VaR
Value at Risk is the
minimum loss that would
be expected a certain
percentage of time over a
certain period of time
given the assumed market
conditions.
In Percentage terms
5% Monthly VaR is 2.8%

In Dollar Terms or Amount

Example 5% Monthly VaR is Rs. 100,000x2.8% = Rs. 2800 – 5%

Implies:
5% probability that the minimum loss is Rs. 2800 or 2.8% in a
given month
95% probability that maximum loss is Rs. 2800 in a given month
• In a 100 month data, if the loss greater than
2.8% / Rs. 2800 occurs more than 5 times it
implies underestimation of VaR
Back-testing
to check if
VaR is right • That implies Actual returns distributions
have fatter tails compared to Normal
Distribution
Estimating Value at Risk – Historical Simulation
Method

• Collect data from historical


lookback period
• Sort data from largest loss to
greatest gain
• Choose the % based on the
chosen confidence intervals
Estimating Value at Risk
– Parametric Method /
Variance –Covariance
Method

• Collect data from historical


lookback period
• Find the Mean & Standard
Deviation
• Standardize the Normal
Distribution
Portfolio Value = Rs. 100000

Monthly Return (mean) = 2%

Example Standard deviation monthly = 3%

Assumption: Normal Distribution


Calculate VaR
Example Cont...

• 10 % Monthly VaR
• Z value = 1.28
• VaR (%) = 2% - (3% x 1.28) = 1.84%
• VaR (amount) = 2% - (3% x 1.28) x 100000 = Rs. 1840

• 5 % Monthly VaR
• Z value = 1.65
• VaR (%) = 2% - (3% x 1.65) = 2.95%
• VaR (amount) = 2% - (3% x 1.28) x 100000 = Rs. 2950
• 1 % Monthly VaR
• Z value = 2.33
Example • VaR (%) = 2% - (3% x 2.33) = 4.99%
• VaR (amount) = 2% - (3% x 2.33) x 100000
Cont... = Rs. 4990
No assumption of Normal Distribution

Monte
Carlo
Useful for large no of assets and risk factors

Simulation Run 1000's of simulations to get the statistical


characteristics
Method Instead of using historical data, generate a
random number that will be used to estimate the
return
Monte Carol
Simulation
Method

You might also like