Matched samples logistic regression in case-control studies with missing values: when to break the matches

Hansson, Lisbeth; Khamis, Harry J.
December 2008
Statistical Methods in Medical Research;Dec2008, Vol. 17 Issue 6, p595
Academic Journal
Simulated data sets are used to evaluate conditional and unconditional maximum likelihood estimation in an individual case-control design with continuous covariates when there are different rates of excluded cases and different levels of other design parameters. The effectiveness of the estimation procedures is measured by method bias, variance of the estimators, root mean square error (RMSE) for logistic regression and the percentage of explained variation. Conditional estimation leads to higher RMSE than unconditional estimation in the presence of missing observations, especially for 1:1 matching. The RMSE is higher for the smaller stratum size, especially for the 1:1 matching. The percentage of explained variation appears to be insensitive to missing data, but is generally higher for the conditional estimation than for the unconditional estimation. It is particularly good for the 1:2 matching design. For minimizing RMSE, a high matching ratio is recommended; in this case, conditional and unconditional logistic regression models yield comparable levels of effectiveness. For maximizing the percentage of explained variation, the 1:2 matching design with the conditional logistic regression model is recommended.


Related Articles

  • Imputation.  // Retail Trade;Dec2009, Vol. 81 Issue 12, p62 

    The article offers information on the process of imputation for the Market Retail Trade Survey (MRTS) in Canada. It notes that imputation is the process used to allot replacement values for missing data. It adds that imputation is established on historical data or administrative data. It...

  • Imputación de ingresos en la Gran Encuesta Integrada de Hogares (GEIH) de 2010. Restrepo Estrada, María Isabel; Marín Diazaraque, Juan Miguel // Desarrollo y Sociedad;2012, Issue 70, p219 

    The aim of this paper is to present the issue in managing surveys with missing data. In order to address this problem, it is reviewed a technique known as imputation. Some methodologies on the income imputation in the 2010 Great Integrated Household Survey (GEIH 2010) were implemented. Seven...

  • Bias in the study of prediction of change: a Monte Carlo simulation study of the effects of selective attrition and inappropriate modeling of regression toward the mean. Gustavson, Kristin; Borren, Ingrid // BMC Medical Research Methodology;2014, Vol. 14 Issue 1, p1 

    Background Medical researchers often use longitudinal observational studies to examine how risk factors predict change in health over time. Selective attrition and inappropriate modeling of regression toward the mean (RTM) are two potential sources of bias in such studies. Method The current...

  • Adjusted Ratio and Regression Type Estimators for Estimation of Population Mean when some Observations are missing. Nangsue, Nuanpan // Proceedings of World Academy of Science: Engineering & Technolog;May2009, Vol. 53, p787 

    Ratio and regression type estimators have been used by previous authors to estimate a population mean for the principal variable from samples in which both auxiliary x and principal y variable data are available. However, missing data are a common problem in statistical analyses with real data....

  • Evaluating a sequential tree-based procedure for multivariate imputation of complex missing data structures. Borgoni, Riccardo; Berrington, Ann // Quality & Quantity;Jun2013, Vol. 47 Issue 4, p1991 

    Item nonresponse in survey data can pose significant problems for social scientists carrying out statistical modeling using a large number of explanatory variables. A number of imputation methods exist but many only deal with univariate imputation, or relatively simple cases of multivariate...

  • A conditional approach for inference in multivariate age-period-cohort models. Held, Leonhard; Riebler, Andrea // Statistical Methods in Medical Research;Aug2012, Vol. 21 Issue 4, p311 

    Age-period-cohort (APC) models are used to analyse data from disease registers given by age and time. When data are stratified by one further variable, for example geographical location, multivariate APC (MAPC) models can be applied to identify and estimate heterogeneous time trends across the...

  • The Case of the Missing Data: Methods of Dealing With Dropouts and Other Research Vagaries. Streiner, David L. // Canadian Journal of Psychiatry;Feb2002, Vol. 47 Issue 1, p68 

    Missing data are common in most studies, especially when subjects are followed over time. This can jeopardize the validity of a study because of reduced power to detect differences, and especially because subjects who are lost to follow-up rarely represent the group as a whole. There are several...

  • TELBS robust linear regression method. Tabatabai, M. A.; Eby, W. M.; Li, H.; Bae, S.; Singh, K. P. // Open Access Medical Statistics;2012 Part 2, Vol. 2, p65 

    Ordinary least squares estimates can behave badly when outlines are present. An alternative is to use a robust regression technique that can handle outliers and influential observations. We introduce a new robust estimation method called TELBS robust regression method. We also introduce a new...

  • The contribution of limited-resource countries to the American Society of Clinical Oncology annual meetings. Masmoudi, Amine // International Journal of Clinical Oncology;Oct2009, Vol. 14 Issue 5, p442 

    Limited-resource countries (LRCs) are underrepresented in biomedical research, and data with respect to oncology are lacking. The aim of the present study was to assess the participation of LRCs in the American Society of Clinical Oncology (ASCO) annual meetings. We analyzed the characteristics...

  • Voluntary Survey Completion Among Team Members: Implications of Noncompliance and Missing Data for Multilevel Research. Hirschfeld, Robert R.; Cole, Michael S.; Bernerth, Jeremy B.; Rizzuto, Tracey E. // Journal of Applied Psychology;May2013, Vol. 98 Issue 3, p454 

    We explored whether voluntary survey completion by team members (in aggregate) is predictable from team members' collective evaluations of team-emergent states. In doing so, we reanalyze less-than-complete survey data on 110 teams from a published field study, using so-called traditional and...


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics