Next: Introduction
Asymptotic T to Z and F to Z Statistic Transformations
FMRIB Technical Report TR00MJ1
Mark Jenkinson and Mark Woolrich
Oxford Centre for Functional Magnetic Resonance Imaging of the Brain (FMRIB),
Department of Clinical Neurology, University of Oxford, John Radcliffe Hospital,
Headley Way, Headington, Oxford, UK
Abstract:
This report describes a mathematical approximation to convert from
T to Z and F to Z statistic values. The approximation is asymptotic as it is
becomes more accurate for higher values of T, F and Z. Since these values
correspond to low values of probability, p, the actual transformation
uses log(p) as the intermediary, and so can be calculated with normal C/C++
float variables, without having problems with underflow.
Mark Jenkinson
2004-01-21