STATISTICAL VALIDATION OF SIMULATION MODELS
BASED ON THE CONCEPT OF PIVOTAL QUANTITIES

Konstantin N. Nechvala, Nicholas A. Nechvalb, Edgars K. Vasermanisb, Kristine Roziteb

a) Department of Computer Science
Transport and Telecommunication Institute
Lomonosov Street 1, LV-1019, Riga, Latvia
e-mail: konstan@tsi.lv

b) Department of Mathematical Statistics
University of Latvia
Raina Blvd 19, LV-1050, Riga, Latvia,
e-mail: nechval@junik.lv

KEYWORDS
Simulation Model; Pivotal Quantity; Validation; Testing.

ABSTRACT
Suppose that we desire to validate a multivariate stationary
response simulation model of an observable system, which has
p response variables. This problem is reduced to testing the
equality of the mean vectors for two multivariate normal
populations, based on the concept of pivotal quantities. In other
words, we consider the multivariate Behrens-Fisher problem
that deals with statistical inference concerning the difference
between the mean vectors of two multivariate normal
populations. In this paper we develop some non-Bayesian exact
solutions based on pivotal quantities. The pivotal quantities we
have developed are functions of the sufficient statistics. The use
of the pivotal quantities is discussed and illustrated via
Hotelling’s two-sample T2 test in testing the validity of a
multivariate stationary response simulation model.