Skip to content

Latest commit

 

History

History
49 lines (32 loc) · 1.57 KB

README.md

File metadata and controls

49 lines (32 loc) · 1.57 KB

Research into calibration of financial models

Things to read/learn about before the starting on the project

  • The R programming language and environment. Start with: http://cran.r-project.org/doc/manuals/R-intro.html

  • Papers on calibration of the Heston model to market prices. Search with Google with "Heston Model Calibration" will lead to some good papers

  • A little bit about QuantLib

Approach to work

  • Define an appropriate interface to QuantLib pricing routines:

    • Assume for time being derivatives wrt parameters are not available
    • Will need to define overall derivative contract specification which does not change
    • Then the quantities which allow multiple observations of the market. Traditionally these will be the maturities and strikes of the options
    • Multiple maturities/strikes should be supplied into a single function call for efficiency
    • The output should be price for each maturity/strike combination

    It is up to us to define the level at the interface works, with options being:

    • Via input/output text files
    • Loading QuantLib as a shared library into R, the using it either via a C-interface, Rcpp (RQuantLib) or SWIG
    • Interprocess call (SOAP, or something similar)
  • Implement optimisation of the model parameters wrt the market observations of prices using standard R least-squares algorithm

    • First, compare the results to the algorithm as implemented in QuantLib!