Original article
Empiric comparison of multivariate analytic techniques: Advantages and disadvantages of recursive partitioning analysis

https://doi.org/10.1016/0021-9681(84)90041-9Get rights and content

Abstract

The recent ready availability of recursive partitioning analysis has provided medical investigators with an alternative to the usual parametric multivariate techniques. We believe that recursive partitioning analysis has several distinct advantages: it provides a simple and intuitive method for classifying subjects, it is designed to identify synergistic interactions among factors, it may identify nonlinear relationships with the endpoint, it provides a simple format for constructing homogeneous risk strata, and it considers prior probabilities and penalties for misclassification in its variable selection process. Conversely, parametric multivariate techniques have several advantages over recursive partitioning analysis: they may be better at finding predictive factors during the later stages of the selection process, they may be less subject to the problems of multiple testing and over-training, and they may be better able to account for the full predictive power of a continuous factor. Although neither type of technique is better in all situations, we believe that recursive partitioning analysis will often be the preferred multivariate method and is especially useful for identifying interaction terms that may then be included in parametric multivariate analyses.

References (31)

  • JA Sonquist et al.

    The detection of interaction effects: A report on a computer program for the selection of optimal combinations of explanatory variables

  • JN Morgan et al.

    THAID: A sequential search program for the analysis of nominal scale dependent variables

    (1973)
  • GV Kass

    An exploratory technique for investigating large quantities of categorical data

    Appl Stat

    (1980)
  • JL Fleiss et al.

    Three computer diagnosis methods compared

    Arch Gen Psychiat

    (1977)
  • RL Spitzer et al.

    DIAGNO: A computer program for psychiatric diagnosis utilizing the differential diagnostic procedure

    Arch Gen Psychiat

    (1968)
  • Cited by (145)

    • Admission Triage With Pain, Inspiratory Effort, Cough Score can Predict Critical Care Utilization and Length of Stay in Isolated Chest Wall Injury

      2022, Journal of Surgical Research
      Citation Excerpt :

      Compared to other multivariable methods, recursive partitioning has advantages and disadvantages. This method generates more accurate, clinically intuitive models that do not require the user to perform calculations and allows varying prioritizing of misclassifications to create a decision rule that has more sensitivity or specificity.27,38,39 However, recursive partitioning does not work well for continuous variables and may overfit data.40

    • Simultaneous determination of phenazopyridine HCl and trimethoprim in presence of phenazopyridine HCl impurity by univariate and multivariate spectrophotometric methods - Quantification of phenazopyridine HCl impurity by univariate methods

      2020, Spectrochimica Acta - Part A: Molecular and Biomolecular Spectroscopy
      Citation Excerpt :

      Univariate methods are characterized by high versatility, sensitivity, and precision. Recently, quantitative spectroscopy has been greatly improved by the use of variety of multivariate statistical methods because the simultaneous inclusion of multiple spectral intensities can greatly improve the precision and applicability of quantitative spectral analysis [28,29]. Multivariate methods have the advantage of small relative prediction errors obtained by smaller number of latent variables, which verify its ability to get accurate results.

    View all citing articles on Scopus

    Supported in part by a grant from the James Hilton Manning and Emma Austin Manning Foundation. Dr Goldman is a Henry J. Kaiser Family Foundation Faculty Scholar in General Internal Medicine.

    View full text