Write a matlab program for backtesting (using actual data FTSE100) the following Value-at-Risk models using the Kupiec(1995) test for coverage and the Christofferson (1998) test for conditional coverage and a method for backtesting ETL due to McNeil and Frey(200) using the confidence levels for VAR/ETL of 99.0% 99.5% and 99.9%.
On page 18 of the article the methodology is clearly stated. The result should appear as those in table 2 to 10 of the article using the following value at risk models
Simple historical simulation
Volatility weighted historical simulation
Parametric normal
Extreme value theory
Parametric student t
Conditional student t
Age weighted historical simulationFor example
Backtest of simple historical simulation
Confidence level Unconditional coverage VaR
P-value Conditional Coverage VaR
p-value. ETL
P-value 1 day horizon Long FTSE100 99.0% 99.5% 99.9% 3 day horizon Long FTSE100 99.0% 99.5% 99.9%
Write a matlab program for backtesting (using actual data FTSE100) the following Value-at-Risk models using the Kupiec(1995) test for coverage and the Christofferson (1998) test for conditional coverage and a method for backtesting ETL due to McNeil and Frey(200) using the confidence levels for VAR/ETL of 99.0% 99.5% and 99.9%.
On page 18 of the article the methodology is clearly stated. The result should appear as those in table 2 to 10 of the article using the following value at risk models
Simple historical simulation
Volatility weighted historical simulation
Parametric normal
Extreme value theory
Parametric student t
Conditional student t
Age weighted historical simulationFor example
Backtest of simple historical simulation
Confidence level Unconditional coverage VaR
P-value Conditional Coverage VaR
p-value. ETL
P-value 1 day horizon Long FTSE100 99.0% 99.5% 99.9% 3 day horizon Long FTSE100 99.0% 99.5% 99.9%
Attachments: