Journal of Statistical Theory and Practice

, Volume 4, Issue 1, pp 169–179 | Cite as

Linear Model Evaluation Based on Estimation of Model Bias

  • Andrew A. NeathEmail author
  • Joseph E. Cavanaugh


Model selection criteria often arise by constructing estimators of measures known as expected overall discrepancies. Such measures provide an evaluation of a candidate model by quantifying the disparity between the true model which generated the observed data and the candidate model. However, attention is seldom paid to the problem of accounting for discrepancy estimator variability, or to the companion problem of establishing discrepancy estimators with certain optimality properties. The expected overall Gauss (error sum of squares) discrepancy for a linear model can be decomposed into a term representing the estimation error, due to unknown model coefficients, and a term representing the approximation error, or bias, due to model misspecification. Since the first error term is seen to depend only on model dimension, a known quantity, the problem of estimating the expected overall Gauss discrepancy reduces to the problem of estimating a bias parameter. In this paper, we derive estimators of model bias with frequentist optimality properties and consider how confidence interval estimation can be used to quantify the uncertainty inherent to the problem of bias parameter estimation. We also show how the problem of estimating model bias can be approached from a Bayesian perspective. To illustrate our methodology, we present a modeling application based on data from a cardiac rehabilitation program at The University of Iowa Hospitals and Clinics.

AMS Subject Classification



Gauss discrepancy Mallows’ Cp Model selection 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Christensen, R., 2002. Plane Answers to Complex Questions. 3rd edition, Springer, New York.CrossRefGoogle Scholar
  2. Clyde, M., George, E.I., 2004. Model uncertainty. Statistical Science, 19, 81–94.MathSciNetCrossRefGoogle Scholar
  3. Davies, S.L., Neath, A.A., Cavanaugh, J.E., 2006. Estimation optimality of corrected AIC and modified Cp in linear regression. International Statistical Review, 74, 161–168.CrossRefGoogle Scholar
  4. Fujikoshi, Y., Satoh, K., 1997. Modified AIC and Cp in multivariate linear regression. Biometrika, 84, 707–716.MathSciNetCrossRefGoogle Scholar
  5. Gelman, A., Carlin, J.B., Stern, H.S., Rubin, D.B., 2003. Bayesian Data Analysis. 2nd edition, Chapman and Hall, London.zbMATHGoogle Scholar
  6. George, E.I., 2000. The variable selection problem. Journal of the American Statistical Association, 95, 1304–1308.MathSciNetCrossRefGoogle Scholar
  7. Gilmour, S.G., 1996. The interpretation of Mallows’ Cp statistic. The Statistician, 45, 49–56.CrossRefGoogle Scholar
  8. Linhart, H., Zucchini, W., 1986. Model Selection. Wiley, New York.zbMATHGoogle Scholar
  9. Mallows, C.L., 1973. Some comments on Cp. Technometrics, 15, 661–675.zbMATHGoogle Scholar
  10. McQuarrie, A.D.R., Tsai, C.-L., 1998. Regression and Time Series Model Selection. World Scientific, River Edge, New Jersey.CrossRefGoogle Scholar

Copyright information

© Grace Scientific Publishing 2010

Authors and Affiliations

  1. 1.Department of Mathematics and StatisticsSouthern Illinois UniversityEdwardsvilleUSA
  2. 2.Department of BiostatisticsThe University of IowaUSA

Personalised recommendations