Welcome back.

In this post, I’ll report quickly on the QuantLib user meeting that was held on July 12th in London thanks to the sponsorship of Quaternion.

However, let me take a couple of news items out of the way first. Goutham and I published an update to the QuantLib Python Cookbook; Goutham added a couple of new notebooks on the Heston model and on caps and floors, while I completed the notebook on Euribor curve bootstrapping and added last post as an appendix. Then, Goutham also announced a project of his that aims at providing a new interface to interact with QuantLib-Python. Do have a look. It’s quite interesting.

So, the user meeting. It was interesting as usual, and the slides are available from the QuantLib site, so you can have a look for yourself. It began with a quick introduction by Roland Lichters, after which Ferdinando Ametrano and I were up. I’ve already told you about the paper we presented, so I’ll shut up now.

Next came a surprise. Roland Lichters, Peter Caspers and Niall O’Sullivan announced Open Risk Engine; that is, an open-source platform for risk and XVA based on QuantLib. The first release will be in September and will be available from GitHub. Details are in the slides.

Then, Eric Ehlers followed with a report on the status of his reposit project, which came a long way since the last time I had a look. It’s now almost to the point where it can be used to generate the existing QuantLibXL functionality.

The last talk of the morning should have been given by Michael von den Driesch, but unfortunately his flight was cancelled. Instead, Sebastian Schlenkrich presented his talk on multi-curve convexity originally scheduled for the afternoon. Besides his slides, you can also read the paper on SSRN on which the presentation was based.

After lunch, another bunch of talks. First, Alexander Sokol reported on the new features in the next release of QuantLibAdjoint and suggested a few future improvements for QuantLib that would make it easier to use AAD.

Then, Daniel Aziz presented his paper (available on SSRN) on a framework for forecasting initial margin requirements.

Finally, Andres Hernandez brought the meeting to an end with a talk on calibration using neural networks; the code he used to obtain the results in his slides is available from GitHub.

As usual, it was a nice mix of financial and technological content, with some interesting bits for everyone. And we might be in for an encore in Düsseldorf in a few months.

Follow me on Twitter if you want to be notified of new posts, or add me to your Google+ circles, or subscribe via RSS: the buttons for that are in the footer. Also, make sure to check my Training page.