FRM Part I · FRM Exam Part I · Simulation and Bootstrapping
In a Monte Carlo simulation of a portfolio's one-day loss, which statement about antithetic variates is correct?
Antithetic variates pair each random draw with its negative and average the two outcomes. For payoffs that move monotonically with the draw, the two results are negatively correlated, so the average has lower variance than two independent draws.
- AThey pair each random draw Z with its negative -Z and average the two results, which reduces variance when the payoff is monotonic in Z.Correct
- BThey replace random draws with a deterministic grid, so the standard error becomes exactly zero.
- CThey increase the number of independent trials by generating draws from a second distribution.
- DThey reduce bias in the estimator by correcting for discretization error in the time steps.
Explanation
Antithetic variates use Z and -Z, creating negatively correlated outputs for monotonic payoffs; averaging them lowers variance. They do not remove sampling error, do not add independent trials, and do not address discretization bias.
Did you get it right without looking?
One question tells you little. A timed set on Simulation and Bootstrapping shows your real accuracy, how long you take and where you lose marks.
More Simulation and Bootstrapping questions
- Which statement best describes a limitation shared by both Monte Carlo simulation and bootstrapping when used for risk estimation?
- A risk analyst estimates the price of a European call by Monte Carlo simulation. For each random draw Z, she also uses -Z to generate a seco…
- A risk team estimates the price of an option by Monte Carlo simulation with N independent draws, obtaining a sample standard deviation of di…
- An analyst estimates the expected payoff of a derivative by Monte Carlo simulation. The standard deviation of the simulated payoffs is 8.0 a…
- A risk manager uses antithetic variates when simulating a stock price with standard normal draws. For each draw Z that generates a payoff X(…
- In a simulation, the standard error of a plain Monte Carlo estimate with 10,000 independent draws is 0.50. The analyst wants the standard er…