Consulting »

Over 15 years financial engineering experience in providing full-range solutions for financial modeling needs.

more…

Time Series Store & Probability Engine »

Sitmo Time Series Store is a high performance and flexible system for managing time series data that has seamless integration with large range of systems.

The Sitmo Probability Engine is a forecasting and Monte Carlo tool for learning complex dynamical behavior from data. It’s currently being used for

• Electricity, Gas and Weather Forward Curves
• load-forecasting
• Interest Rate, and Volatility forward cures and behaviour

Popular Articles

Some Experiments with our Non-Linear Denoising Algorithm

June 5th, 2012

We have been doing some benchmark-tests of our novel denoising and forecasting algorithm. The algorithm is based an extension of the reconstruction theory of Takens to stochastic delay differential equations, and is extr[...]

Light Intensity Modelling For Electricity Forward Curves

April 5th, 2012

Light intensity modelling is one the most important but also most often flawed element in electricity demand-, generation- and price-modelling. In this article we present the detailed properties of the our high quality l[...]

The Distribution of The Sample Correlation

March 12th, 2012

When you generate a set of correlated random number and then calculate the correlation of those samples, you'll notice that the calculated "sample correlation" can deviate from the correlation value use to generate the [...]

Generating Correlated Random Numbers

February 29th, 2012

This article describes common methods that are used in generating correlated random numbers. [latexpage] Generating two sequences of correlated random numbers Generating two sequences of random numbers with a [...]

An Internally Consistent Interpolation Method for Yield Curves

November 11th, 2011

Here we present a new yield curve interpolation method, one that's based on conditioning a stochastic model on a set of market yields. The concept is closely related to a Brownian bridge where you generate scenario accor[...]

Spread Option Pricing Model

June 3rd, 2011

This equation uses the Gauss-Legendre quadrature to approximate the value of a spread option. The Gauss-Legendre quadrature abscissas (Xi) are rescaled in the range -4 to +4. The equation is unbiased and gives very accu[...]

Calculating Correlation and Means with Missing Data

June 3rd, 2011

A great deal has been written about fixing 'invalid correlation matrices' for risk management purposes. Correlation matrices are invalid when it's mathematically impossible to generate random numbers with those mutual c[...]

A Simple and Extremely Fast C++ Template for Matrices and Tensors

May 31st, 2011

This article describes a very simple and efficient solution for handling matrices and tensors in C++. The idea is to store the matrix (or tensor) in a standard vector by translating the multidimensional index to a one d[...]