Archive for January, 2007

Cholesky decomposition & Correlated Normals

Wednesday, January 31st, 2007

I was talking to a friend of mine the other day, about, financial science of course, and the question came up: “I know that by transforming a vector of uncorrelated normals with the cholesky-factoring the correlation matrix will result in a vector of correlated normals… but why does that work?”. I was going to rig up the argument in a hurry, in a bar, on the back of a napkin, but I couldn’t recall it. I promised my friend I’d give him a very simple one-liner argument and the intention here is to fulfill the promise.

First, recall that the correlation matrix of a vector  \overrightarrow{y} of unit normals is given by

\displaystyle \Big\langle \overrightarrow{y} \cdot \overrightarrow{y}^\textrm{T} \Big\rangle = \left( \begin{array}{cccc}1 & \rho_{12} & \cdots & \rho_{1N} \\ \rho_{21} & 1 & \vdots & \vdots \\ \vdots & \cdots & \ddots & \rho_{N-1,N} \\ \rho_{N1} & \cdots & \rho_{N,N-1} & 1 \\ \end{array} \right) = \rho

(Notice by the way, that there is an easy fix to the stupid 038 appearing in the matrix, fix explained here but for some reason I don’t have write access to the file /wp-content/plugins/latexrender/latex.php so we’ll have to wait for this to be legible.)

Anyway, that was just to recall what the correlation matrix was. Now assume we Cholesky factorize it as follows:

\displaystyle C^\textrm{T} \cdot C = \rho

and consider the outcome of an uncorrelated vector normal transformed by the cholesky factor:

\displaystyle \overrightarrow{z} = C \cdot \overrightarrow{x}

We know that the this must be a unit vector normal, the only thing left to determine is its correlation matrix:

\displaystyle \Big\langle \overrightarrow{z} \cdot \overrightarrow{z}^\textrm{T} \Big\rangle  = \Big\langle C \cdot \overrightarrow{x} \cdot \overrightarrow{x}^\textrm{T} \cdot C^\textrm{T} \Big\rangle

But the cholesky matrix is deterministic and comes out of the expectation, and using the uncorrelated property of x we get:

\displaystyle \Big\langle \overrightarrow{z} \cdot \overrightarrow{z}^\textrm{T} \Big\rangle  = C \cdot \Big\langle \overrightarrow{x} \cdot \overrightarrow{x}^\textrm{T} \Big\rangle \cdot C^\textrm{T} = C \cdot C^\textrm{T} = \rho^\textrm{T} = \rho

And since the only thing which prevented the last two equations to fit in one line is the unusually narrow margins here, the argument has been presented as a one-liner.

LaTeX Renderer Is Here

Thursday, January 25th, 2007

Our beloved admin has honored us with the installation of the LaTeX Renderer… oh yeah baby!! Now we can actually blog in mathematics! Check this out for example:

 \textrm{d} X_t = \mu X_t \, \textrm{d} t + \sigma X_t \, \textrm{d} W_t.

There we go with the Black-Scholes SDE model, right here, with LaTeX fonts in a blog. Anyways, I’m gonna start off with some structured content right now, this is supposed to only be a newsflash.

Sælir IceQuants félagar

Wednesday, January 10th, 2007

Sælir allir IceQuants félagar!! Þótt það séu kannski ekki lengur neinar nýjar fréttir þá eru það engu að síður fréttir að ég er hættur hjá Landsbankanum. Þið vitið þetta allir, enda erum við ekki margir meðlimirnir hér. Ég vona nafið á mínum vinnustað hafi engin áhrif á þessa IceQuants pælingu okkar, enda átti þetta að vera “inter-institutional” frá upphafi.

Mig langar að gera smá tilraun með það hversu margir eru að lesa hér og hverjir eru aktívir notendur. Hver sem sér þetta entry verður strax að gera comment á það, segja bara “hæ, ég las þetta” svo við getum séð hversu langan tíma það tekur þar til allir hafa svarað.

Svo er það annað… ég keypti nýju útgáfuna af “Arbitrage Theory in Continuous Time” eftir Tomas Björk. Þessi bók er “by-the-way” snilld. Nýja útgáfan hefur að geyma talsvert dýpri meðferð á stochastic processum og málfræði fyrir þá sem fíla það, en samt er hægt að lesa hana í gegn án þess (hann merkir erfiða kafla með stjörnu). Mæli eindregið með þessari. Ég keypti líka ágætis undergrad texta sem fjallar um stoch. prósessa. Ég keypti hana vegna þess að hún tekur á þessum Poisson prósessum (flestar tala bara um Weiner prósessa). Ég kem með gagnrýni á þessa bók þegar ég er búinn að lesa meira í henni.

Thats all folks.

Skúli