Summarizing the predictive power of a generalized linear model
- PMID: 10861777
- DOI: 10.1002/1097-0258(20000715)19:13<1771::aid-sim485>3.0.co;2-p
Summarizing the predictive power of a generalized linear model
Abstract
This paper studies summary measures of the predictive power of a generalized linear model, paying special attention to a generalization of the multiple correlation coefficient from ordinary linear regression. The population value is the correlation between the response and its conditional expectation given the predictors, and the sample value is the correlation between the observed response and the model predicted value. We compare four estimators of the measure in terms of bias, mean squared error and behaviour in the presence of overparameterization. The sample estimator and a jack-knife estimator usually behave adequately, but a cross-validation estimator has a large negative bias with large mean squared error. One can use bootstrap methods to construct confidence intervals for the population value of the correlation measure and to estimate the degree to which a model selection procedure may provide an overly optimistic measure of the actual predictive power.
Copyright 2000 John Wiley & Sons, Ltd.
Similar articles
-
Bias reduction in logistic dose-response models.J Biopharm Stat. 2011 May;21(3):405-22. doi: 10.1080/10543401003703306. J Biopharm Stat. 2011. PMID: 21442516
-
Estimating the expectation of the log-likelihood with censored data for estimator selection.Lifetime Data Anal. 2004 Dec;10(4):351-67. doi: 10.1007/s10985-004-4772-z. Lifetime Data Anal. 2004. PMID: 15690990
-
A comparison of methods to handle skew distributed cost variables in the analysis of the resource consumption in schizophrenia treatment.J Ment Health Policy Econ. 2002 Mar;5(1):21-31. J Ment Health Policy Econ. 2002. PMID: 12529567
-
Variance and confidence limits in validation studies based on comparison between three different types of measurements.J Epidemiol Biostat. 2000;5(5):303-13. J Epidemiol Biostat. 2000. PMID: 11142606
-
Segmented regression with errors in predictors: semi-parametric and parametric methods.Stat Med. 1997 Jan 15-Feb 15;16(1-3):169-88. doi: 10.1002/(sici)1097-0258(19970130)16:2<169::aid-sim478>3.0.co;2-m. Stat Med. 1997. PMID: 9004390
Cited by
-
Analysing monkeypox epidemic drivers: Policy simulation and multi-index modelling across 39 nations.J Glob Health. 2024 Feb 9;14:04037. doi: 10.7189/jogh.14.04037. J Glob Health. 2024. PMID: 38333932 Free PMC article.
-
Excessive fear of clusters of holes, its interaction with stressful life events and the association with anxiety and depressive symptoms: large epidemiological study of young people in Hong Kong.BJPsych Open. 2023 Aug 14;9(5):e151. doi: 10.1192/bjo.2023.540. BJPsych Open. 2023. PMID: 37577835 Free PMC article.
-
MRI radiomics for brain metastasis sub-pathology classification from non-small cell lung cancer: a machine learning, multicenter study.Phys Eng Sci Med. 2023 Sep;46(3):1309-1320. doi: 10.1007/s13246-023-01300-0. Epub 2023 Jul 17. Phys Eng Sci Med. 2023. PMID: 37460894
-
What is the impact of national public expenditure and its allocation on neonatal and child mortality? A machine learning analysis.BMC Public Health. 2023 Apr 28;23(1):793. doi: 10.1186/s12889-023-15683-y. BMC Public Health. 2023. PMID: 37118765 Free PMC article.
-
On the Relative Importance of Different Factors Explaining Health Plan Choices: Evidence From Mandatory Health Insurance in Switzerland.Front Health Serv. 2022 Apr 15;2:847486. doi: 10.3389/frhs.2022.847486. eCollection 2022. Front Health Serv. 2022. PMID: 36925810 Free PMC article.
Publication types
MeSH terms
Grants and funding
LinkOut - more resources
Other Literature Sources