High frequency trading agent based model calibration using sequential Monte Carlo approximate Bayesian Computation : Software
softwareposted on 04.09.2020 by Kelly Goosen, Tim Gebbie
Code as a research output can either be uploaded directly from your computer or through the code management system GitHub. Versioning of code repositories is supported.
Python code to aid reproducibility of the dissertation: Calibrating High Frequency Trading Data to Agent Based Models using Approximate Bayesian Computation. Detailed steps for reproducibility can be found in the README.md document. The code includes several functions and scripts. Included are also the resulting plots for the paper. The research focuses on replicating the Preis-Golke-Paul-Schneider Agent Based Model (Preis et al., 2006) and attempting to calibrate this model along with a simple auto-regressive moving average model using sequential Monte Carlo approximate Bayesian computation. DOI for the dataset used: http://dx.doi.org/10.17632/nt8nw28h7c.1.