1,493 research outputs found
Pairwise parameter estimation in Rasch models
No full textRasch model item parameters can be estimated
consistently with a pseudo-likelihood method based
on comparing responses to pairs of items irrespective
of other items. The pseudo-likelihood method is
comparable to Fischer’s (1974) Minchi method. A
simulation study found that the pseudo-likelihood
estimates and their (estimated) standard errors were
comparable to conditional and marginal maximum
likelihood estimates. The method is extended to
estimate parameters of the linear logistic test model
allowing the design matrix to vary between persons.
Index terms: item parameter estimation, linear logistic
test model, Minchi estimation, pseudo-likelihood,
Rasch model.Zwinderman, Aeilko H.. (1995). Pairwise parameter estimation in Rasch models. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/118398
Variance ratios for each of the three datasets, under Box-Cox transformation, for different values of λ
No full text<p><b>Copyright information:</b></p><p>Taken from "Comparing transformation methods for DNA microarray data"</p><p>BMC Bioinformatics 2004;5():77-77.</p><p>Published online 17 Jun 2004</p><p>PMCID:PMC449698.</p><p>Copyright © 2004 Thygesen and Zwinderman; licensee BioMed Central Ltd. This is an Open Access article: verbatim copying and redistribution of this article are permitted in all media for any purpose, provided this notice is preserved along with the article's original URL.</p
Variance ratio for the Box-Cox transform as a function of λ, for 10 complementary subsets of the genes
No full text<p><b>Copyright information:</b></p><p>Taken from "Comparing transformation methods for DNA microarray data"</p><p>BMC Bioinformatics 2004;5():77-77.</p><p>Published online 17 Jun 2004</p><p>PMCID:PMC449698.</p><p>Copyright © 2004 Thygesen and Zwinderman; licensee BioMed Central Ltd. This is an Open Access article: verbatim copying and redistribution of this article are permitted in all media for any purpose, provided this notice is preserved along with the article's original URL.</p> Testis data
Robustness of marginal maximum likelihood estimation in the Rasch model
No full textSimulation studies examined the effect of misspecification
of the latent ability (θ) distribution on the accuracy
and efficiency of marginal maximum likelihood
(MML) item parameter estimates and on MML statistics
to test sufficiency and conditional independence. Results
were compared to the conditional maximum likelihood
(CML) approach. Results showed that if θ is assumed
to be normally distributed when its distribution
is actually skewed, MML estimators lose accuracy and
efficiency when compared to CML estimators. The effects
are not large, though they increase as the skewness
of the number-correct score distribution increases.
However, statistics to test the sufficiency and conditional
independence assumptions of the Rasch model
in the MML approach are very sensitive to misspecification
of the θ distribution. Index terms: ability distribution,
conditional likelihood, efficiency, goodness
of fit, marginal likelihood, Rasch model, robustness.Zwinderman, Aeilko H.; Van den Wollenberg, Arnold L.. (1990). Robustness of marginal maximum likelihood estimation in the Rasch model. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/107788
A generalized rasch model for manifest predictors
No full textlatent trait, Rasch model, logistic regression, conditional marginal likelihood,
Members of a highly widespread bacteriophage family are hallmarks of metabolic syndrome gut microbiomes
No full textThere is significant interest in altering the course of cardiometabolic disease development via the gut microbiome. Nevertheless, the highly abundant phage members -which impact gut bacteria- of the complex gut ecosystem remain understudied. Here, we characterized gut phageome changes associated with metabolic syndrome (MetS), a highly prevalent clinical condition preceding cardiometabolic disease. MetS gut phageome populations exhibited decreased richness and diversity, but larger inter-individual variation. These populations were enriched in phages infecting Bacteroidaceae and depleted in those infecting Ruminococcaeae. Differential abundance analysis identified eighteen viral clusters (VCs) as significantly associated with either MetS or healthy phageomes. Among these are a MetS-associated Roseburia VC that is related to healthy control-associated Faecalibacterium and Oscillibacter VCs. Further analysis of these VCs revealed the Candidatus Heliusviridae, a highly widespread gut phage lineage found in 90+% of the participants. The identification of the temperate Ca. Heliusviridae provides a novel starting point to a better understanding of the effect that phages have on their bacterial hosts and the role that this plays in MetS
Associating multiple longitudinal traits with high-dimensional single-nucleotide polymorphism data: application to the Framingham Heart Study
No full textABSTRACT : Cardiovascular diseases are associated with combinations of phenotypic traits, which are in turn caused by a combination of environmental and genetic factors. Because of the diversity of pathways that may lead to cardiovascular diseases, we examined the so-called intermediate phenotypes, which are often repeatedly measured. We developed a penalized nonlinear canonical correlation analysis to associate multiple repeatedly measured traits with high-dimensional single-nucleotide polymorphism dat
Machine Learning in Medicine: A Complete Overview
No full textThe amount of data stored in the world’s databases doubles every 20 months, as estimated by Usama Fayyad, one of the founders of machine learning and co-author of the book Advances in Knowledge Discovery and Data Mining (ed. by the American Association for Arti? cial Intelligence, Menlo Park, CA, USA, 1996), and clinicians, familiar with traditional statistical methods, are at a loss to analyze them. T raditional methods have, indeed, dif? culty to identify outliers in large datasets, and to ? nd patterns in big data and data with multiple exposure/outcome variables. In addition, analysis-rules for surveys and questionnaires, which are currently common methods of data collection, are, essentially, missing. Fortunately, the new discipline, machine learning, is able to cover all of these limitations. S o far, medical professionals have been rather reluctant to use machine learning. Ravinda Khattree, co-author of the book Computational Methods in Biomedical Research (ed. by Chapman & Hall, Baton Rouge, LA, USA, 2007) suggests that there may be historical reasons: technological (doctors are better than computers (?)), legal, cultural (doctors are better trusted). Also, in the ? eld of diagnosis making, few doctors may want a computer checking them, are interested in collaboration with a computer or with computer engineer
- …
