I am willing to do my Master Thesis about option pricing. Do you have any suggestions? I would like it to be something simple, like comparing methods, e.g. compare ARCH and GARCH approaches for volatility estimates and their impact options pricing . Or this could be some empiriacl study using data from the Market. Any suggestions of something simple? Please try to detail a little bit because I am a beginner in the subject - I've fully read Hull and I'm starting to read Schreve. Thanks!

  • $\begingroup$ Frankly, just by reading Hull's is not yet ready for a thesis on option pricing, unless you already have a good background in probability, statistics, or econometrics. $\endgroup$ – Gordon Dec 14 '16 at 15:13
  • $\begingroup$ @Gordon yes, I have a good background in statistics. Would you have any suggestion of a subject for my master thesis? $\endgroup$ – John Dec 14 '16 at 16:00
  • $\begingroup$ I believe you need to discuss with your supervisor. Most of us are industry professionals. $\endgroup$ – Gordon Dec 14 '16 at 16:08

In option pricing, the entire game is fitting the skew with a fairly robust model. All the research right now is in LSV (Local Stochastic Vol) Models. Fitting these is a challenge (with PDE or Particle Methods), maybe a study on that will be ideal if you're looking for a derivatives job after.

Alternately, you could also test ML techniques in obtaining the vol surface (or even estimating vol models)

  • $\begingroup$ Why do you say fitting these is a challenge? Jaeckel's hyphyp model (with a few adjustments to make it more flexible - ie time dependent params) can be diffused as a 2d pde for fitting, and works very nicely... $\endgroup$ – will Jan 25 '17 at 23:09
  • $\begingroup$ @Drew What is ML? $\endgroup$ – John Jan 30 '17 at 22:14
  • $\begingroup$ Machine Learning $\endgroup$ – Drew Feb 2 '17 at 0:23
  1. Fit GARCH to data.
  2. Use obtained GARCH model and Monte Carlo simulation method for pricing options.
  3. Compare speed of classical and quasi-Monte Carlo algorithms.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.