Wald-Wolfowitz Test of Randomness
The function wawotest.default performs a Wald Wolfowitz test of
the random distribution of the values in a vector. The function
wawotest.ltraj performs this tests for the descriptive
parameters dx, dy and dist in an object of class
ltraj. The function wawotest is generic.
wawotest(x, ...)
## Default S3 method:
wawotest(x, alter = c("greater", "less"), ...)
## S3 method for class 'ltraj'
wawotest(x, ...)x |
for |
alter |
a character string specifying the alternative hypothesis, must be one of "greater" (default), "less" or "two-sided" |
... |
additional arguments to be passed to other functions |
The statistic of the test is equal to A = sum(y(i) y(i+1)), with y(N+1) = y(1). Under the hypothesis of a random distribution of the values in the vector, this statistic is normally distributed, with theoretical means and variances given in Wald & Wolfowitz (1943).
wawotest.default returns a vector containing the value of the
statistic (a), its esperance (ea), its variance
(va), the normed statistic (za) and the
P-value. wawotest.ltraj returns a table giving these values for
the descriptive parameters of the trajectory.
Stephane Dray dray@biomserv.univ-lyon1.fr
Wald, A. & Wolfowitz, J. (1943) An exact test for randomness in the non-parametric case based on serial correlation. Ann. Math. Statist., 14, 378–388.
indmove and runsNAltraj for other
tests of independence to be used with objects of class "ltraj"
data(puechcirc) puechcirc wawotest(puechcirc)
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.