Calibration and Documentation

This repository holds diagnostic and research for calibrating option prices using the empirical characteristic function. The main contributions are:

• The use of the empirical characteristic function for use in parametric option pricing calibration
• The use of genetic and meta-heurstic algorithms in overcoming the highly non-linear optimization problem

However, after several experiments, it appears that using "traditional" L-BFGS directly on the option prices (not transforming into the complex domain) obtains superior results.

Requirements

The documentation is written in R Sweave. The application is written in Rust. To efficiently generate the json files needed for the documentation, use Node.

Steps to run

• Clone this repo and cd into the folder
• cargo build --release
• node index
• Open OptionCalibration in a Sweave/Latex editor (eg RStudio) and compile.

Relevant links

The difficult work is done by some of my dependent Crates:

• fang_oost
• fang_oost_option
• cuckoo
• cf_functions