Premium
Some monte carlo estimates of the yule distribution
Author(s) -
Simon Herbert A.,
Van Wormer Theodore A.
Publication year - 2007
Publication title -
behavioral science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.371
H-Index - 45
eISSN - 1099-1743
pISSN - 0005-7940
DOI - 10.1002/bs.3830080305
Subject(s) - monte carlo method , zipf's law , statistical physics , class (philosophy) , computer science , mathematics , statistics , artificial intelligence , physics
The pioneering work of Zipf in establishing statistical regularities of verbal output has spurred on the construction of stochastic models of language behavior of which the observed regularities would be the deduced consequences. A class of such models is subsumed under socalled Yule processes. Only special cases of such processes have been solved exactly. More general cases have yielded only approximate solutions. The Monte Carlo method gives an idea of how good these approximations are and also provides an opportunity to compare verbal output data with simulated Yule processes directly.