Intra-cluster correlation structure in longitudinal data analysis: Selection criteria and misspecification tests

Selection criteria and misspecification tests for the intra-cluster correlation structure (ICS) in longitudinal data analysis are considered. In particular, the asymptotical distribution of the correlation information criterion (CIC) is derived and a new method for selecting a working ICS is proposed by standardizing the selection criterion as the p-value. The CIC test is found to be powerful in detecting misspecification of the working ICS structures, while with respect to the working ICS selection, the standardized CIC test is also shown to have satisfactory performance. Some simulation studies and applications to two real longitudinal datasets are made to illustrate how these criteria and tests might be useful.

Modeling strategies in longitudinal data analysis: Covariate, variance function and correlation structure selection

A modeling paradigm is proposed for covariate, variance and working correlation structure selection for longitudinal data analysis. Appropriate selection of covariates is pertinent to correct variance modeling and selecting the appropriate covariates and variance function is vital to correlation structure selection. This leads to a stepwise model selection procedure that deploys a combination of different model selection criteria. Although these criteria find a common theoretical root based on approximating the Kullback-Leibler distance, they are designed to address different aspects of model selection and have different merits and limitations. For example, the extended quasi-likelihood information criterion (EQIC) with a covariance penalty performs well for covariate selection even when the working variance function is misspecified, but EQIC contains little information on correlation structures. The proposed model selection strategies are outlined and a Monte Carlo assessment of their finite sample properties is reported. Two longitudinal studies are used for illustration.

Working-correlation-structure identification in generalized estimating equations

Selecting an appropriate working correlation structure is pertinent to clustered data analysis using generalized estimating equations (GEE) because an inappropriate choice will lead to inefficient parameter estimation. We investigate the well-known criterion of QIC for selecting a working correlation Structure. and have found that performance of the QIC is deteriorated by a term that is theoretically independent of the correlation structures but has to be estimated with an error. This leads LIS to propose a correlation information criterion (CIC) that substantially improves the QIC performance. Extensive simulation studies indicate that the CIC has remarkable improvement in selecting the correct correlation structures. We also illustrate our findings using a data set from the Madras Longitudinal Schizophrenia Study.

Computerized synthesis of the structure of geared kinematic chains

The paper presents for the first time a fully computerized method for structural synthesis of geared kinematic chains which can be used to derive epicyclic gear drives. The method has been formulated on the basis of representing these chains by their graphs, the graphs being in turn represented algebraically by their vertex-vertex incidence matrices. It has thus been possible to make advantageous use of concepts and results from graph theory to develop a method amenable for implementation on a digital computer. The computerized method has been applied to the structural synthesis of single-freedom geared kinematic chains with up to four gear pairs, and the results obtained thereform are presented and discussed.

Structure Of 5'-Deoxy-5',6-Epithio-5,6-Dihydro-2',3'-O-Isopropylideneuridine, C12h16n2o5s

Mr=300.33 , triclinic, P1, a=5.635 (2), b=11.077(2), c=11.582(2)A, a= 70.48 (1), fl= 88.16 (3), y=80.56(3) ° , V= 670.325 A3, Z=2, D x = 1.49 Mg m -3, Cu Ka, n= 1.54184 ,A, g = 2.308mm -1, F(000)=316, T=301K, R=0.054, R w = 0.093 for 1944 observed counter reflections. The sulphur position with respect to the dihydrouracil ring, which is of possible relevance to the action of thymidylate synthetase, is axial in molecule A and equatorial in B. Both molecules show the anti conformation about the glycosidic bond [torsion angle C(6)-N(1)-C(1')-O(4'), 2'CN = 21.6 (9) and 29.4 (10) °] and have the C(4')-endo, O(4')-exo (40T) sugar conformation. The dioxolane-ring conformation is O(2')-endo in A and C(7)-endo in B. The dihydrouracil rings show self base pairing with hydrogen bondsN(3A)...O(ZB) and N(3B)...O(ZA).

The Phraseology and Structure of Latin Building Inscriptions in Roman North Africa

This study analyses the diction of Latin building inscriptions. Despite its importance, this topic has rarely been discussed before: the most substantial contribution on the subject is a short dissertation by Klaus Gast (1965) that focuses on 100 inscriptions dating mostly from the Republican period. Marietta Horster (2001) also touched upon this theme in her thesis on imperial building inscriptions. I have collected my source material in North Africa because more Latin building inscriptions dating from the Imperial period have survived there than in any other area of the Roman Empire. By means of a thorough and independent survey, I have assembled all relevant African Latin building inscriptions datable to the Roman period (between 146 BC and AD 425), 1002 texts, into a corpus. These inscriptions are all fully edited in Appendix 1; Appendix 2 contains references to earlier editions. To facilitate search operations, both are also available in electronic form. They are downloadable from the address http://www.helsinki.fi/hum/kla/htm/jatkoopinnot.htm. Chapter one is an introduction dealing with the nature of building inscriptions as source material. Chapter two offers a statistical overview of the material. The following main section of the work falls into five chapters, each of which analyses one main part of a building inscription. An average building inscription can be divided into five parts: the starting phrase opens the inscription (a dedication to gods, for example), the subject part identifies the builder, the object part describes the constructed or repaired building, the predicate part records the building activity and the supplement part offers additional information on the project (it can specify the funding, for instance). These chapters are systematic and chronological and their purpose is to register and interpret the phrases used, to analyse reasons for their use and for their popularity among the different groups of builders. Chapter eight, which follows the main section of the work, creates a typology of building inscriptions based on their structure. It also presents the most frequently attested types of building inscriptions. The conclusion describes, on a general level, how the diction of building inscriptions developed during the period of study and how this striking development resulted from socio-economic changes that took place in Romano-African society during Antiquity. This study shows that the phraseology of building inscriptions had a clear correlation both with the type of builder and with the date of carving. Private builders tended to accentuate their participation (especially its financial side) in the project; honouring the emperor received more emphasis in the building inscriptions set up by communities; the texts produced by the army were concise. The chronological development is so clear that it enables stylistic dating. At the beginning of the imperial period the phrases were clear, concrete, formal and stereotyped but by Late Antiquity they have become vague, subjective, flexible, varied and even rhetorically or poetically coloured.

A polarized, twisted, ethylene: structure of methyl 2-(1,3-dimethyl-2-imidazolidinylidene)-3-oxobutyrate dihydrate, C10H16N2O3.2H2O

Mr = 248, monoclinic, P21/n, a = 12.028 (2), b=7.168(2), c= 15.187(5)A, fl=91.88(2) °, Z= 4, V= 1308.6,~3, Din= 1.26, Dx= 1.263 Mgm -3, 2 (Cu Ka) = 1.5418 .A, g = 0.86 mm -1, F(000) = 536, T= 293 K. Final R = 5.6% for 2120 observed reflexions. Owing to the push-pull effect, the C=C bond distance is as long as 1.464 (2)/k with the twist angle about the bond 62.6.

Morphological structure of bismuth-doped n-type amorphous germanium sulphide semiconductors

Recent observation of n-type conduction in amorphous Ge20Ss_xBix at large bismuth concentrations (x = 11), which otherwise shows p-type conduction, has aroused considerable interest in the international scientific community [1]. The mechanism of such impurity incorporation in a germanium chalcogenide glass is not understood and is a topic of current interest. In our recent publications [2-10] we have brought to light some hitherto unknown and interesting features of bismuth dopants in chalcogen-rich Ge-X (X -- S, Se) glassy compositions. In this communication we present our new results of investigations on vitreous semiconductors Ge20S80 Bi using electron microscopy, electron diffraction of as-prepared and annealed/pressure quenched compositions. Our results provide conclusive support to the formation of composite clusters containing all the three elements, germanium, sulphur and bismuth, which crystallize in simpler stoichiometric compounds Bi2S3 and GeS2.

Structure of a thionucleoside: 5'-deoxy-5',6-epithio-5,6-dihydro-2',3'-O-isopropylidene-3-methyluridine

C13HlsN205 S, M r = 314.35, orthorhombic, P212121 with a = 39.526 (4), b = 6.607 (2), c = 5.661 (2) A, Z = 4, V = 1478.36 A 3, D c = 1.412 Mg m -3, Cu Ka radiation. Final R = 0.073 for 1154 observed counter reflections. The sulphur atom is in a pseudo-equatorial position with respect to the dihydrouracil ring. The sugar pucker is predominantly O(l')-exo unlike the C(3')-exo,C(4')-endo observed for 2',3'-O-isopropylideneuridine (ISPU). The fivemembered dioxolane ring has C(7) displaced by 0.497 (7)A from the best plane through atoms 0(2'), C(2'), C(3'), 0(3'), in contrast to ISPU where 0(3') shows the maximum deviation.

Deep inelastic electroproduction structure functions for polarized and unpolarized proton targets

Following Ioffe's method of QCD sum rules the structure functions F2(x) for deep inelastic ep and en scattering are calculated. Valence u-quark and d-quark distributions are obtained in the range 0.1 less, approximate x <0.4 and compared with data. In the case of polarized targets the structure function g1(x) and the asymmetry Image Full-size image are calculated. The latter is in satisfactory agreement in sign and magnitude with experiments for x in the range 0.1< x < 0.4.

Criteria for working-correlation-structure selection in GEE: Assessment via simulation

Efficiency of analysis using generalized estimation equations is enhanced when intracluster correlation structure is accurately modeled. We compare two existing criteria (a quasi-likelihood information criterion, and the Rotnitzky-Jewell criterion) to identify the true correlation structure via simulations with Gaussian or binomial response, covariates varying at cluster or observation level, and exchangeable or AR(l) intracluster correlation structure. Rotnitzky and Jewell's approach performs better when the true intracluster correlation structure is exchangeable, while the quasi-likelihood criteria performs better for an AR(l) structure.

Population structure, mortality and growth of Pinna nobilis Linnaeus, 1758 (Mollusca, Bivalvia) at different depths in Moraira bay (Alicante, Western Mediterranean)

An investigation to characterize the causes of Pinna nobilis population structure in Moraira bay (Western Mediterranean) was developed. Individuals of two areas of the same Posidonia meadow, located at different depths (A1, -13 and A2, -6 m), were inventoried, tagged, their positions accurately recorded and monitored from July 1997 to July 2002. On each area, different aspects of population demography were studied (i.e. spatial distribution, size structure, displacement evidences, mortality, growth and shell orientation). A comparison between both groups of individuals was carried out, finding important differences between them. In A1, the individuals were more aggregated and mean and maximum size were higher (A1, 10.3 and A2, 6 individuals/100 m(2); A1, x = 47.2 +/- 9.9; A2, x = 29.8 +/- 7.4 cm, P < 0.001, respectively). In A2, growth rate and mortality were higher, the latter concentrated on the largest individuals, in contrast to A1, where the smallest individuals had the higher mortality rate [A1, L = 56.03(1 - e(-0.17t)); A2, L = 37.59(1 - e(-0.40t)), P < 0.001; mean annual mortality A1: 32 dead individuals out of 135, 23.7% and A2: 16 dead individuals out of 36, 44.4%, and total mortality coefficients (z), z(A1(-30)) = 0.28, z(A1(31-45)) = 0.05, z(A1(46-)) = 0.08; z(A2(-30)) = 0.15, z(A2(31-45)) = 0.25]. A common shell orientation N-S, coincident with the maximum shore exposure, was observed in A2. Spatial distribution in both areas showed not enough evidence to discard a random distribution of the individuals, despite the greater aggregation on the deeper area (A1) (A1, chi(2) = 0.41, df = 3, P > 0.5, A2, chi(2)= 0.98, df = 2 and 0.3 < P < 0.5). The obtained results have demonstrated that the depth-related size segregation usually shown by P. nobilis is mainly caused by differences in mortality and growth among individuals located at different depths, rather than by the active displacement of individuals previously reported in the literature. Furthermore, dwarf individuals are observed in shallower levels and as a consequence, the relationship between size and age are not comparable even among groups of individuals inhabiting the same meadow at different depths. The final causes of the differences on mortality and growth are also discussed.

Working correlation structure misspecification, estimation and covariate design: Implications for generalised estimating equations performance

The method of generalised estimating equations for regression modelling of clustered outcomes allows for specification of a working matrix that is intended to approximate the true correlation matrix of the observations. We investigate the asymptotic relative efficiency of the generalised estimating equation for the mean parameters when the correlation parameters are estimated by various methods. The asymptotic relative efficiency depends on three-features of the analysis, namely (i) the discrepancy between the working correlation structure and the unobservable true correlation structure, (ii) the method by which the correlation parameters are estimated and (iii) the 'design', by which we refer to both the structures of the predictor matrices within clusters and distribution of cluster sizes. Analytical and numerical studies of realistic data-analysis scenarios show that choice of working covariance model has a substantial impact on regression estimator efficiency. Protection against avoidable loss of efficiency associated with covariance misspecification is obtained when a 'Gaussian estimation' pseudolikelihood procedure is used with an AR(1) structure.