Selection of important variables and determination of functional form for continuous predictors in multivariable model building
Top Cited Papers
- 3 December 2007
- journal article
- research article
- Published by Wiley in Statistics in Medicine
- Vol. 26 (30), 5512-5528
- https://doi.org/10.1002/sim.3148
Abstract
In developing regression models, data analysts are often faced with many predictor variables that may influence an outcome variable. After more than half a century of research, the ‘best’ way of selecting a multivariable model is still unresolved. It is generally agreed that subject matter knowledge, when available, should guide model building. However, such knowledge is often limited, and data-dependent model building is required. We limit the scope of the modelling exercise to selecting important predictors and choosing interpretable and transportable functions for continuous predictors. Assuming linear functions, stepwise selection and all-subset strategies are discussed; the key tuning parameters are the nominal P-value for testing a variable for inclusion and the penalty for model complexity, respectively. We argue that stepwise procedures perform better than a literature-based assessment would suggest. Concerning selection of functional form for continuous predictors, the principal competitors are fractional polynomial functions and various types of spline techniques. We note that a rigorous selection strategy known as multivariable fractional polynomials (MFP) has been developed. No spline-based procedure for simultaneously selecting variables and functional forms has found wide acceptance. Results of FP and spline modelling are compared in two data sets. It is shown that spline modelling, while extremely flexible, can generate fitted curves with uninterpretable ‘wiggles’, particularly when automatic methods for choosing the smoothness are employed. We give general recommendations to practitioners for carrying out variable and function selection. While acknowledging that further research is needed, we argue why MFP is our preferred approach for multivariable model building with continuous covariates. Copyright © 2007 John Wiley & Sons, Ltd.Keywords
This publication has 26 references indexed in Scilit:
- Generalized Additive ModelsPublished by Taylor & Francis Ltd ,2006
- Predictors of mortality and morbidity in patients with chronic heart failureEuropean Heart Journal, 2005
- Dichotomizing continuous predictors in multiple regression: a bad ideaStatistics in Medicine, 2005
- Building Multivariable Regression Models with Continuous Covariates in Clinical EpidemiologyMethods of Information in Medicine, 2005
- Confessions of a pragmatic statisticianJournal of the Royal Statistical Society: Series D (The Statistician), 2002
- Regression Modeling StrategiesPublished by Springer Science and Business Media LLC ,2001
- Building Multivariable Prognostic and Diagnostic Models: Transformation of the Predictors by Using Fractional PolynomialsJournal of the Royal Statistical Society Series A: Statistics in Society, 1999
- Flexible smoothing with B-splines and penaltiesStatistical Science, 1996
- Commentary: Prognostic models: clinically useful or quickly forgotten?BMJ, 1995
- Regression Using Fractional Polynomials of Continuous Covariates: Parsimonious Parametric ModellingJournal of the Royal Statistical Society Series C: Applied Statistics, 1994