I met with one of my professors today. For some background, he is a professor in the mathematics department who studies statistics (specifically kernel density estimation and nonparametric curve estimation). We were talking about other professor’s research, and I mentioned I was interested in stochastic calculus and mathematical finance. He looks at me and says, quietly so no one else can here, “It doesn’t work. You can’t accurately model financial behavior; it just doesn’t work.”

This really got me thinking. Why did he say this? Surely stochastic processes, like Brownian motion, play a heavy role in mathematical finance. Unfortunately, it made me second guess studying anything like this. Any thoughts?

Thanks in advance.

  • 1
    $\begingroup$ Maybe he can't accurately model financial behaviour. The successful hedge funds clearly can. $\endgroup$ – Patrick Stevens Oct 6 '18 at 8:01
  • 1
    $\begingroup$ Do hedge funds still use Black Scholes and stochastic calculus techniques? Or have they all switched to machine learning or something instead. $\endgroup$ – littleO Oct 6 '18 at 8:04
  • $\begingroup$ @littleO I feel like data science and machine learning techniques are dominating right now-pretty much drowning out whatever is left of studying the usual probability models. $\endgroup$ – MathIsLife12 Oct 6 '18 at 8:17
  • $\begingroup$ @littleO There's a distinction between modeling to interpolate prices of illiquid stuff from prices of liquid stuff (and to optimally hedge said illiquid stuff with liquid stuff) and modeling to predict future prices and volatility. $\endgroup$ – spaceisdarkgreen Oct 6 '18 at 8:24
  • 1
    $\begingroup$ @PatrickStevens or they earn their money from something else entirely and it's all a facade.. $\endgroup$ – mathreadler Oct 6 '18 at 10:20

Your Answer

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

Browse other questions tagged or ask your own question.