Nematrian
/

Market Consistency and WMC [24]

Go to: Summary | Previous | Next   
Bullet points include: Usually Monte Carlo involves equally weighted simulations. For e.g. the problem of integrating an unknown function, using equal weights typically minimises variance (hence error) of result. Although may use e.g. stratified sampling, relevant sample points then focus on mathematically ‘important’ part of the problem, e.g. where the integrand is large. But it is possible to use non-equal weights. See e.g. Avellaneda et.al. (2001). “Weighted Monte Carlo: A new technique for calibrating asset-pricing models” or Kemp (2009)

NAVIGATION LINKS
Contents | Prev | Next | Library

Desktop view | Switch to Mobile