Historical simulation method: Difference between revisions
imported>Doug Williamson (Expand.) |
imported>Doug Williamson (Layout.) |
||
Line 1: | Line 1: | ||
1. | '''1.''' | ||
In Value at Risk analysis, an alternative to the Delta-normal method of calculating the underlying probability distribution. | In Value at Risk analysis, an alternative to the Delta-normal method of calculating the underlying probability distribution. | ||
Historical simulation is conceptually the simplest alternative method to the delta-normal. There is no assumption about how markets operate. | |||
For any given portfolio held today, you calculate repeatedly its hypothetical value change as if it had been held for a one day period in the past, using the relevant market price changes and other market rate changes for each successive day. | For any given portfolio held today, you calculate repeatedly its hypothetical value change as if it had been held for a one day period in the past, using the relevant market price changes and other market rate changes for each successive day. | ||
Line 9: | Line 10: | ||
At each step, you do a full valuation and calculate the ex-post or historical value changes over one day. | At each step, you do a full valuation and calculate the ex-post or historical value changes over one day. | ||
Finally, tabulate the empirical distribution of one-day value changes and identify the | Finally, tabulate the empirical distribution of one-day value changes and identify the adverse 95% point. This point is the basis of the one-day 95% VaR. | ||
2. | '''2.''' | ||
Similar methods in other risk analysis applications. | Similar methods in other risk analysis applications. |
Revision as of 15:10, 30 October 2016
1.
In Value at Risk analysis, an alternative to the Delta-normal method of calculating the underlying probability distribution.
Historical simulation is conceptually the simplest alternative method to the delta-normal. There is no assumption about how markets operate.
For any given portfolio held today, you calculate repeatedly its hypothetical value change as if it had been held for a one day period in the past, using the relevant market price changes and other market rate changes for each successive day.
At each step, you do a full valuation and calculate the ex-post or historical value changes over one day.
Finally, tabulate the empirical distribution of one-day value changes and identify the adverse 95% point. This point is the basis of the one-day 95% VaR.
2.
Similar methods in other risk analysis applications.