v1.8.0

A new AE algorithm was implemented: the Bayesian QAE one. The BAYESQAE class from QQuantLib.AE.bayesian_ae modules implements the algorithm. This class can be included in other workflows in a transparent way (like the quantum integration or the amplitude estimation for pricing options).

1.8.0

Minor addition to README

v1.7.0

This new release allows the user to configure price estimation problems for Cliquet options a more complicated financial derivative whose final payoff can be positive or negative.

The cliquet_return_estimation module from QQuantLib.finance package configures the Cliquet option problem, the Amplitude Estimation algorithm and a properly configured QPU. The price problem is solved and a pandas DataFrame with all the information is returned. In this case, the payoff is loaded completely into the quantum state (using square or direct encoding) so, only the RQAE algorithms will provide a correct solution if the final price is negative.

The cliquet_return_estimation_step_payoff from QQuantLib.finance package works like the previous model but the payoff is loaded by parts (positive and negative) and then post-process the output to provide the correct estimation. In this case, traditional AE algorithms will provide a correct solution if the final price is negative.

In the benchmark/q_ae_cliquet/ folder we provide the modules benchmark_cliquet and benchmark_cliquet_step_po. These scripts execute, from the command line, compute the price of a Cliquet option using a QAE algorithm. The configuration of the Cliquet, the AE algorithm and the QPU can be provided using JSON files. Several JSON file examples can be found in the folder. Additionally, the QAE_CliquetOptions.ipynb jupyter notebook that serves as the tutorial is placed inside it.

v1.6.1

Update setup.py

1.6.1

Update setup.py

NEASQC final project release

The Financial Applications software library and the associated QQuantLib Python library gathers all the myqlm code implementations relative to the use case Financial applications of the WP 5 of the NEASQC project.

This release corresponds to the end of the project one.
The QQuantLib library develops new contributions to two main areas critical for bank industry:

• Quantum Accelerated Monte Carlo (QAMC) for option pricing (corresponding to the packages DL, AA, PE, AE, finance from QQuantLib )
• Quantum Machine Learning (QML) for risk assessment (corresponding to the package qml4var from QQuantLib )

The benchmark package of the Financial Applications software library gathers all the different benchmarks utilities related with the major contributions developed under the framework of the project:

• new encoding algorithm for loading negative payoffs
• new Amplitude Estimation (AE) algorithm that has a similar performance than other state-of-the-art AE algorithms
• new Cost Function that improves the behaviour of PQCs for using them as surrogate models for risk assessment

v1.0

Release 1.0 of the Financial Application library.