/stochastic-processes-kyotoUx009x

Solutions and extensions to "KyotoUx-009x: Stochastic Processes: Data Analysis and Computer Simulation"

Primary LanguageJupyter Notebook

stochastic-processes-kyotoUx009x

Solutions and extensions to "KyotoUx-009x: Stochastic Processes: Data Analysis and Computer Simulation".

This repositry houses code that I extended and developed for the course. It features, among others:

  • Simulations of stochastic processes

    • Random Walk
    • Browinian Motion, Diffusion (also with particle interactions, with drift)
    • Ornstein-Uhlenbeck Process
    • The Heston Stochastic Volatility Model, Geometric Brownian Motion
    • Stochastic Dealer Model (also with memory/drift)

  • Numerical methods (Forward/Backward Euler, Runge-Kutta 2nd/4th, Leap-Frog)

  • Analysis of stochastic processes

    • (auto)correlation, spectral density (FFT)
    • Maxwel-Boltzmann distribution, Green-Kubo formula
    • absolute price return, logarithmic change in price
    • real world stock data
    • fitting distribution to data

  • Visualizations

    • Code for producing animations in jupyter notebook
    • nbagg backend, ggplot style

I highly recommend this course to anybody who seeks well structured and approachable introduction to stochastic processes. You should have basic idea about various numerical methods and have some experience with python programming before you start. You needn't to be any expert though, this course will help you develop your skills also in these two aspects.