Skip to content

FRM Part I · FRM Exam Part I · Simulation and Bootstrapping

Compared with a Monte Carlo simulation that assumes returns follow a normal distribution, what is the principal advantage of the bootstrap for estimating the distribution of a statistic?

The bootstrap's main advantage is that it needs no assumed parametric distribution; it resamples the observed data and so preserves features like fat tails and skewness. It cannot create unobserved extremes and depends on the sample being representative.

  1. AIt does not require a specified parametric distribution because it resamples the empirical data, so it can capture fat tails present in the sampleCorrect
  2. BIt eliminates sampling error entirely because it uses the full population
  3. CIt guarantees correct results even when the original sample is very small or unrepresentative
  4. DIt generates data outside the range of observed values, so it captures extreme events not seen historically

Explanation

The bootstrap is nonparametric: it draws from observed data, retaining features such as skewness and fat tails without assuming a distribution. It cannot produce values beyond the observed range and is only as good as the original sample, so it does not remove sampling error.

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.

Practise Simulation and BootstrappingFree diagnostic mock

More Simulation and Bootstrapping questions