Table of Contents

relu - calculate the relative uncertainty statistic (ERP statistics)
**relu ** [ **-s** ] *base* *maxlag*
**relu ** calculates the relative
uncertainty statistic. The user specifies *maxlag*, the maximum desired
lag, and the program will calculate the U value for that lag and all smaller
lags. The user must also specify the *base* of the data. The data are expected
to be single digit integers between 0 and *base*-1, and it is assumed that
*base* equals the number of possible values (e.g. if there were two possible
values, they would be coded as 0 and 1, and the *base* would be 2). The data
are read from the standard input, and don’t have to be in any special kind
of format. The individual data can be separated by any non-numeric characters
(usually spaces, tabs, or newlines), but they don’t need to be separated
by anything at all. It is therefore possible to have data that look like
this:

01100100101001111101
10010001110100110101010
101 Non-numeric characters
11 like these are ignored!!!!
101011
001010101111010011011001011000110101111001010001

The **-s** option can be used to produce a shorter output that is appropriate
for building files filled with U values. For example,

relu -s 2 10 < datafile >> outputfile

would be appropriate for data in base 2, with 10 different lags. The data
would be read from a file named datafile and the output would be appended
to the end of a file named outputfile.

chisquar(1)
, freqdist(1)
,
mean(1)
, median(1)
, mnsdse(1)
, zscore(1)
Steve Luck

**Table of Contents**