# how to represent financial data as a spatial process

Does any one have a good tutorial , introduction or overview on the web for different ways of representing financial data as a spatial process? Such as those spatial processes often used in geo-statistics.

EDIT

What I am looking for is going from financial time series data to fit the following general framework for spatial processes

let $s \in \Re$ be a generic data location in $d$-dimensional Euclidian space and suppose the potential datnum $Z(s)$ at spatial location $s$ is a random quantity. Let $D$ be a random set. Now let $s$ vary over index $D \subset \Re^d$ so as to generate a multivariate field or process:

$\qquad \qquad \qquad \qquad \qquad \qquad \qquad \{ Z(s):\in D \};$

This may also be more commonly referred to as a random field

-
I've been pondering over a similar idea and believe for my application representing the data in a spatial network makes sense. My main objective is visualization - ie remove the temporal component to highlight meaningful spatial relationships in the data. Within a network representation this can be easily done by coloring the connections between nodes wrt their weights for instances. Not sure if that's what you're after with your question though. – emsfeld Feb 21 at 2:31
Can you give an example? emsfeld is considering networks relating quantities, but I thought at first you meant viewing a stochastic process as a random walk. So an example would give us some handle on the question. – Phil H Feb 21 at 8:19
@Phil H - yes after writing my comment and rereading the question I was a bit confused as well. I may have misinterpreted the question and was more thinking along the lines of visualizing orderbook related quantities within a network representation, so my comment may be a bit unrelated to the question by pyCthon. – emsfeld Feb 21 at 8:44
Will update with an example shortly – pyCthon Feb 22 at 2:43