Generative Modeling for Time Series Via Schrödinger Bridge

This repo is the official code for our paper Generative Modeling for Time Series Via Schrödinger Bridge, available at https://papers.ssrn.com/abstract_id=4412434

Quickstart

Python

To get started, create a conda environment and install the required Python packages for this project using the following command:

conda create -name SBTimeSeries --file requirements.txt python=3.8.16

Build C++ Code

Windows

To build the code on Windows, open the Visual Studio solution SBTimeSeries.sln and compile it.

Linux

For Linux users, you can build the solution by running the batch file Build.sh using the following command:

bash Build.sh

Repository Structure

The repository is organized as follows:

• src directory contains the C++ code for SBTS diffusion..
• deepHedging directory contains the TensorFlow model for deep hedging and its data generator.
• notebook contains two Jupyter Notebooks files: SBTSNumericalExperiments.ipynb which allows you to reproduce all the numerical experiments presented in the paper and generates samples stored in a folder named data (created automatically) for deep hedging. DeepHedging.ipynb uses the generated samples to run deep hedging as described in the paper.