PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 11764257

Local influence to detect influential data structures for generalized linear mixed models.

M J Ouwens·F E Tan·M P Berger
View original source ↗Export metadata ↓

Abstract

This article discusses the generalization of the local influence measures for normally distributed responses to local influence measures for generalized linear models with random effects. For these models, it is shown that the subject-oriented influence measure is a special case of the proposed observation-oriented influence measure. A two-step diagnostic procedure is proposed. The first step is to search for influential subjects. A search for influential observations is proposed as the second step. An illustration of a two-treatment, multiple-period crossover trial demonstrates the practical importance of the detection of influential observations in addition to the detection of influential subjects.

Explore related subjects

AspartameBiometryClinical Trials as TopicCross-Over StudiesHeadacheHumansLinear ModelsJournal Article

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

M J Ouwens, F E Tan, M P Berger. 2001. Local influence to detect influential data structures for generalized linear mixed models.. https://doi.org/10.1111/j.0006-341x.2001.01166.x

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2001

Kinetics of enzymatic solid-to-solid peptide synthesis: synthesis of Z-aspartame and control of acid-base conditions by using inorganic salts.

Enzymatic peptide synthesis can be carried out efficiently in solid-to-solid reaction mixtures with 10% (w/w) water added to a mixture of substrates. The final reaction mass contains >/=80% (by weight) of product. This article deals with acid-base effects in such reaction mixtures and the consequences for the enzyme. In the Thermoase-catalyzed synthesis of Z-Asp-Phe-OMe, the reaction rate is strongly dependent on the amount of basic salts added to the system. The rate increases 20 times, as the KHCO(3) or K(2)CO(3) added is raised 2.25-fold from an amount equimolar to the Phe-OMe. HCL starting material. With further increases in KHCO(3) addition, the initial rate remains at the maximum, but with K(2)CO(3) it drops sharply. Addition of NaHCO(3) is less effective, but rates are faster if more water is used. With >1.5 equivalents of basic salt, the final yield of the reaction decreases. Similar effects are observed when thermolysin catalyzes the same reaction, or Z-Gln-Leu-NH(2) synthesis. These effects can be rationalized using a model estimating the pH of these systems, taking into account the possible formation of up to ten different solid phases.

Aspartame↗
PubMed2001

Structure, dynamics, and stability of beta-cyclodextrin inclusion complexes of aspartame and neotame.

Studies of the high-intensity sweetener aspartame show that its stability is significantly enhanced in the presence of beta-cyclodextrin (beta-CyD). At a 5:1 beta-CyD/aspartame molar ratio, the stability of aspartame is 42% greater in 4 mM phosphate buffer (pH 3.1) compared to solutions prepared without beta-CyD. Solution-state (1)H NMR experiments demonstrate the formation of 1:1 beta-CyD/aspartame complexes, stabilized by the interaction of the phenyl-ring protons of aspartame with the H3 and H5 protons of beta-CyD. Inclusion complex formation clearly accounts for the observed stability enhancement of aspartame in solution. The formation of inclusion complexes in solution is also demonstrated for beta-CyD and neotame, a structural derivative of aspartame containing an N-substituted 3,3-dimethylbutyl group. These complexes are stabilized by the interaction of beta-CyD with both phenyl-ring and dimethylbutyl protons. Solid-state NMR experiments provide additional characterization, clearly demonstrating the formation of inclusion complexes in lyophilized solids prepared from solutions of beta-CyD and either aspartame or neotame.

Aspartame↗