Abstract
Statistical models are sometimes incorporated into computer software for making predictions about future observations. When the computer model consists of a single statistical model this corresponds to estimation of a function of the model parameters. This paper is concerned with the case that the computer model implements multiple, individually-estimated statistical sub-models. This case frequently arises, for example, in models for medical decision making that derive parameter information from multiple clinical studies. We develop a method for calculating the posterior mean of a function of the parameter vectors of multiple statistical models that is easy to implement in computer software, has high asymptotic accuracy, and has a computational cost linear in the total number of model parameters. The formula is then used to derive a general result about posterior estimation across multiple models. The utility of the results is illustrated by application to clinical software that estimates the risk of fatal coronary disease in people with diabetes.
Similar content being viewed by others
References
Brown, J.B.: Computer models of diabetes: almost ready for prime time. Diabet. Res. Clin. Pract. 50(3), S1–3 (2000)
Clarke, P.M., Gray, A.M., Briggs, A., Farmer, A.J., Fenn, P., et al.: A model to estimate the lifetime health outcomes of patients with type 2 diabetes: the United Kingdom Prospective Diabetes Study (UKPDS) Outcomes Model (UKPDS no. 68). Diabetologia 47, 1747–1759 (2004)
Eastman, R.C., Javitt, J.C., Herman, W.H., Dasbach, E.J., Zbrozek, A.S., et al.: Model of complications of NIDDM.I. Model construction and assumptions. Diabet. Care 20, 725–734 (1997)
Eddy, D.M., Schlessinger, L.: Archimedes: A trial-validated model of diabetes. Diabet. Care 26, 3093–3101 (2003)
Frost, C., Thompson, S.: Correcting for regression dilution bias: comparison of methods for a single predictor variable. J. Roy. Stat. Soc. Ser. A 163, 173–190 (2000)
Hoeffding, W.: Asymptotic normality. In: Kotz, S., Johnson, N.L., Read, C.B. (eds.) Encyclopedia of Statistical Sciences. Wiley, New York (1982)
Stevens, R.J.: Evaluation of methods for interval estimation of model outputs, with application to survival models. J. Appl. Stat. 30, 967–981 (2003)
Stevens, R.J., Coleman, R.L., et al.: Risk factors for myocardial infarction case fatality and stroke case fatality in type 2 diabetes (UKPDS 66). Diabet. Care 27, 201–207 (2004)
Sweeting, T.J.: Approximate Bayesian computation based on signed roots of log-density ratios (with discussion). In: Bernardo, J.M., Berger, J.O., Dawid, A.P., Smith, A.F.M. (eds.) Bayesian Statistics 5. Oxford University Press, Oxford (1996)
Sweeting, T.J., Kharroubi, S.A.: Some new formulae for posterior expectations and Bartlett corrections. Test 12, 497–521 (2003)
Tierney, L., Kadane, J.B.: Accurate approximations for posterior moments and marginal densities. J. Am. Stat. Assoc. 81, 82–86 (1986)
UKPDS Group: The UKPDS Risk Engine: a model for the risk of coronary heart disease in type 2 diabetes (UKPDS 56). Clin. Sci. 101, 671–679 (2001)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Stevens, R.J., Sweeting, T.J. Estimation across multiple models with application to Bayesian computing and software development. Stat Comput 17, 245–252 (2007). https://doi.org/10.1007/s11222-007-9026-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-007-9026-y