Project Fact File

Title: Unbiased estimation of multifractal dimensions of finite data sets
Category: maths
Area: multiscale spatial structures
No. of units: 1 or 2 or 4
Supervisor: Tony Roberts (Staff Profile)
Description:

Explore a novel method to determine the entire multi-fractal spectrum from an experimental sample of points. The long term aim is to develop a robust method based on analysing the data by comparing it with synthetic multi-fractals and seeking a maximum likelihood match. The method is designed to eliminate major biases inherent in other methods of determining fractal dimensions from limited samples. These techniques may be applied to data on the spatial distribution of underwater plants. For some algae these are proving to have a multi-fractal pattern, at least on a small-scale. Over the last few years I collaborated with Herbert Jelinek on the estimation of multifractal dimensions of cells to help diagnose neural disease.

Student: