::: UL LUSENN :::

Lusenn is a division of Sivienn that specializes in identifying patterns in time series data. It addresses the challenge posed by the overwhelming amount of data available to traditional mathematical models.

Biology, physics, finance, economics and other social sciences are each affected in their own way. Lusenn promotes the formulation and use of probabilities, which have proven their effectiveness in all these areas over time. In finance, stochastic and local volatility models have been developed since the mid-1990s. While they improve market analysis by capturing the effects of implied volatility, they struggle with the multi-scale properties of the data they process. Time series, for instance, exhibit scale variations that complicate the development of models capable of reproducing multiple-scale phenomena. One of Lusenn's goals is to address these challenges by creating models that replicate such complicated processes for use in pricing and hedging derivatives.

Lusenn is part of Sivienn.

Calibration

The models developed by Lusenn's mathematicians become operational as soon as they are calibrated using the available data. The observable parameters are taken into account when creating and refining the models. Lusenn has its own specific estimation methods, which are based on wavelets and multi-scale techniques.

Simulation

Practical applications require Monte-Carlo simulation models. Lusenn has built and implemented specific algorithms that meet these requirements.

Pricing

The pricing of products and derivative contracts is a major application of the models developed by Lusenn. The modeling framework makes it possible to capture the effects of multiple scales for pricing.

Hedging

A price calculation formula is not enough. It must be supplemented by a portfolio risk quantification procedure. Lusenn develops the right procedure, which creates an effective hedging strategy.