Generalized estimating equations for ordinal categorical data: arbitrary patterns of missing responses and missingness in a key covariate

We propose methods for regression analysis of repeatedly measured ordinal categorical data when there is nonmonotone missingness in these responses and when a key covariate is missing depending on observables. The methods use ordinal regression models in conjunction with generalized estimating equations (GEEs). We extend the GEE methodology to accommodate arbitrary patterns of missingness in the responses when this missingness is independent of the unobserved responses. We further extend the methodology to provide correction for possible bias when missingness in knowledge of a key covariate may depend on observables. The approach is illustrated with the analysis of data from a study in diagnostic oncology in which multiple correlated receiver operating characteristic curves are estimated and corrected for possible verification bias when the true disease status is missing depending on observables.     

A model for thermograms

Thermograms are curves resulting from thermal analysis and are of great interest in the study of various food and biological products physical properties. A method to separate underlying peaks is proposed, and statistical properties of estimates for some characteristic parameters are derived. The total number of peaks can be estimated with a sequential analysis of the residual plots. For each new peak, a statistical criterion is proposed to check whether it is significantly different from the noise of the recording. As an example, the method is applied to a summer milk fat fusion thermogram.     

Predictive margins with survey data

In the analysis of covariance, the display of adjusted treatment means allows one to compare mean (treatment) group outcomes controlling for different covariate distributions in the groups. Predictive margins are a generalization of adjusted treatment means to nonlinear models. The predictive margin for group r represents the average predicted response if everyone in the sample had been in group r. This paper discusses the use of predictive margins with complex survey data, where an important consideration is the choice of covariate distribution used to standardize the predictive margin. It is suggested that the textbook formula for the standard error of an adjusted treatment mean from the analysis of covariance may be inappropriate for applications involving survey data. Applications are given using data from the 1992 National Health Interview Survey (NHIS) and the Epidemiologic Followup Study to the first National Health and Nutrition Examination Survey (NHANES I).     

Approximate standard errors in semiparametric models

We consider semiparametric models with p regressor terms and q smooth terms. We obtain an explicit expression for the estimate of the regression coefficients given by the back-fitting algorithm. The calculation of the standard errors of these estimates based on this expression is a considerable computational exercise. We present an alternative, approximate method of calculation that is less demanding. With smoothing splines, the method is exact, while with loess, it gives good estimates of standard errors. We assess the adequacy of our approximation and of another approximation with the help of two examples.     

Bayesian survival analysis using a MARS model

A Bayesian multivariate adaptive regression spline fitting approach is used to model univariate and multivariate survival data with censoring. The possible models contain the proportional hazards model as a subclass and automatically detect departures from this. A reversible jump Markov chain Monte Carlo algorithm is described to obtain the estimate of the hazard function as well as the survival curve.     

Relationship Between Amino Acid Properties and Protein Stability: Buried Mutations

In order to understand the mechanism of protein stability and to develop a simple method for predicting mutation-induced stability changes, we analyzed the relationship between stability changes caused by buried mutations and changes in 48 amino acid properties. As expected from the importance of hydrophobicity, properties reflecting hydrophobicity are strongly correlated with the stability of proteins. We found that subgroup classification based on secondary structure increased correlations significantly, and mutations within -strand segments correlated better than did those in -helical segments, which may result from stronger hydrophobicity of the -strands. Multiple regression analyses incorporating combinations of three properties from among all possible combinations of the 48 properties increased the correlation coefficient to 0.88 and by an average of 13% for all data sets. Analyzing the stability of tryptophan synthase mutants with Glu49 replaced by all other residues except Arg revealed that combining buriedness, solvent-accessible surface area for denatured protein, and unfolding Gibbs free energy change increased the correlation to 0.95. Consideration of sequence and structural information (neighboring residues in sequence and in space) did not significantly strengthen the correlations in buried mutations, suggesting that nonspecific interactions dominate in the interior of proteins.     

Comparison of parameters from rhythmometric models with multiple components on hybrid data

Population multiple components is a statistical tool useful for the analysis of time-dependent hybrid data. With a small number of parameters, it is possible to model and to predict the periodic behavior of a population. In this article, we propose two methods to compare among populations rhythmometric parameters obtained by multiple component analysis. The first is a parametric method based in the usual statistical techniques for comparison of mean vectors in multivariate normal populations. The method, through MANOVA analysis, allows comparison of the MESOR and amplitude-acrophase pair of each component among two or more populations. The second is a nonparametric method, based in bootstrap techniques, to compare parameters from two populations. This test allows one to compare the MESOR, the amplitude, and the acrophase of each fitted component, as well as the global amplitude, orthophase, and bathyphase estimated when all fitted components are harmonics of a fundamental period. The idea is to calculate a confidence interval for the difference of the parameters of interest. If this interval does not contain zero, it can be concluded that the parameters from the two models are different with high probability. An estimation of p-value for the corresponding test can also be calculated. Both methods are illustrated with an example, based on clinical data. The nonparametric test can also be applied to paired data, a special situation of great interest in practice. By the use of similar bootstrap techniques, we illustrate how to construct confidence intervals for any rhythmometric parameter estimated from population multiple components models, including the orthophase, bathyphase, and global amplitude. These tests for comparison of parameters among populations are a needed tool when modeling the nonsinusoidal rhythmic behavior of hybrid data by population multiple component analysis.     

Stand characteristics in colour-infrared aerial photographs as indicators of epiphytic lichens

Information about forest biodiversity has so far been collected mostly by using field inventories, but it is desirable to find other methods that can cover large areas at a lower cost. In a forest landscape covering 2000 ha in southern Sweden we tested if colour-infrared (CIR) aerial photographs on the scale of 1:30000 can be used to interpret forest stand characteristics correlated to the occurrence of epiphytic lichens that are Red-listed or otherwise indicate high nature conservation value (signal species). Using logistic regression we found that the interpreted stand variables tree height and crown structure class were significantly correlated to the occurrence of Red-listed species. For signal species, the variables tree height, percentage of southern deciduous trees and crown structure class were significantly correlated to the occurrence. The logistic regression models successfully predicted a significantly higher probability to find Red-listed species in the stands that actually contained such species compared to stands without Red-listed species. The same was true for stands with signal species. We conclude that interpretation of CIR aerial photographs could be a useful method to find certain groups of epiphytic lichens in surveys covering large areas.     

高血压大鼠心肌肥大及逆转过程中相关因素的探讨

目的：探讨在心肌肥大及逆转过程中收缩压(SBP)、舒张压(DBP)、平均动脉压(MAP)、神经肽Y(NPY)等与左心室肥大的关系。方法：血压和心率用生物信号分析系统记录;NPY用放射免疫法测定,用SPSS软件求出了相关系数和回归方程。结果：SBP、DBP、MAP、心肌匀浆中NPY与心系数(LVW／BW)呈正相关,血液中NPY和心率(HR)与心系数不相关。结论：血压升高是导致左室肥大的因素之一,收缩压的影响大于舒张压;SBP、DBP、MAP、心肌匀浆中NPY与心系数(LVW/BW)有相关的趋势。     

Estimating the length of incomplete long bones: forensic standards from Guatemala

We report on new standards for estimating long bone length from incomplete bones for use in forensic and archaeological contexts in Central America. The measurements we use closely follow those defined by Steele ([1970] Personal Identification in Mass Disasters; Washington, DC: Smithsonian Institution), but we add several new landmarks. We measured the femur, humerus, tibia, and fibula of 100 Maya skeletons (68 males, 32 females) recovered from forensic exhumations. We derived the equations by regressing bone segment length on bone length, and solved for bone length to maximize the utility of the equations for taller populations. We generated equations for all segments that were significantly correlated with bone length for males, for females, and for both sexes combined, but accepted only regressions with r(2) > 0.85 as reliable. Landmarks defined by muscle attachment sites were more variable in location than landmarks on articular architecture; thus we retained few equations that use these landmarks. We tested the male and combined sex equations on 36 males of unknown ethnicity exhumed from a military base in Guatemala, and found that the equations performed satisfactorily. We also evaluated the performance of equations by Steele ([1970] Personal Identification in Mass Disasters; Washington, DC: Smithsonian Institution) and Jacobs ([1992] Am J Phys Anthropol 89:333-345) on the Maya bones, and conclude that significant population variation in long bone proportions hinders their application in Central America.