< Back to Volume 13, Number 1
Volume 13: Number 1: Article 3
A Rescaled Range Analysis of Random Events1
Fotini Pallikari and Emil Boller, Institut für Grenzgebiete der Psychologie,
Wilhelmstrasse 3A, D-79098 Freiburg, Germany
The rescaled range statistical analysis was applied on sets of random
numbers to demonstrate its potential in studying various types of biases
and the presence of periodical features. The data were generated by
electronic random number generators in psychokinesis tests. According
to the theory of Hurst, the rescaled range of independent random data
is proportional to the square root of their number. In data which are
not independent, the fractional Brownian motion model of Mandelbrot
is useful in modeling their time series as persistent or anti-persistent.
A weak predominantly persistent type of fractional Brownian motion in
the data indicated a bias which could not be distinguished from chance
fluctuations after comparison with computer simulated data. The basic
steps for the application of this method, the variety of information
it can provide and its limitations are discussed. The method provides
a relatively simple, yet robust, technique for studying anomalies in
random events.
1 Part of the work was presented at the 1996
Euro SSE conference. Request for reprints should be addressed to: Fontini
Pallikari, Physics Department, University of Athens, Zografos, Panepistimiopolis,
Athens 157 84, Greece.
Keywords: anomalies, fractional Brownian motion, Hurst exponent, periodicities,
psychokinesis, rescaled range analysis
FULL TEXT:
A Rescaled Range Analysis of Random Events
To purchase back issues contact Allen Marketing & Management: 1-800-627-0629
