You are not currently logged in.
Access your personal account or get JSTOR access through your library or other institution:
Regression Analysis for Simulation Practitioners
Jack P. C. Kleijnen
The Journal of the Operational Research Society
Vol. 32, No. 1 (Jan., 1981), pp. 35-43
Stable URL: http://www.jstor.org/stable/2581467
Page Count: 9
Preview not available
Some simple extensions of elementary regression analysis useful for analyzing simulation experiments are given. In simulation variance estimates (standard errors) are usually available but often do not satisfy the assumption of constant variance which underlies elementary regression analysis. Two approaches are possible: either switch to Generalized (or Weighted) Least Squares or continue to use Ordinary Least Squares. The consequences of both approaches are surveyed. Testing the adequacy of the regression model is discussed in detail. A case-study illustrates the statistical techniques. Alternatives to Least Squares are briefly indicated.
The Journal of the Operational Research Society © 1981 Operational Research Society