Nonparametric Two-Sample Tests for General Clinical Data

Starting from the discussion of a practical example a unifying concept for the derivation of meaningfully interpretable nonparametric tests for the two-sample case is developed which may well be adapted for other designs, too. This methodology covers other well-known procedures, e.g. Fisher's exact test, the Wilcoxon-Mann-Whitney and Gehan's tests, and may furthermore be extended to all situations sharing the same fundamental structural property of the sample space, namely its strict order induced by the substantial problem under study. The resulting test procedure is discussed for a randomization argument, exact and approximate, as well as for the general specific test problem. A numerical example is provided.     

Covariate analysis of viral eradication studies

An important issue arising in therapeutic studies of hepatitis C and HIV is the identification of and adjustment for covariates associated with viral eradication and resistance. Analyses of such data are complicated by the fact that eradication is an occult event that is not directly observable, resulting in unique types of censored observations that do not arise in other competing risks settings. This paper proposes a semiparametric regression model to assess the association between multiple covariates and the eradication/resistance processes. The proposed methods are based on a piecewise proportional hazards model that allows parameters to vary between observation times. We illustrate the methods with data from recent hepatitis C clinical trials.     

Plant phenological data and tree-rings as palaeoclimate indicators in south-west Finland since AD 1750

Plant phenological data and tree-rings were tested for their palaeoclimatic value in south-west Finland since AD 1750. The information from fragmentary, partly overlapping, partly non-systematically biased plant phenological records of 14 different phenomena (a total of 3,144 observations) was combined into one continuous time series of phenological indices. All site- and phenomenon-specific series were standardized to present an average of zero and standard deviation of one. The mean phenomenon-specific series were then averaged as arithmetic means for annually resolved time series representing the variability in the particular plant phenomenon. Consequently, each phenomenon-specific mean series was based on spatially normalized site-specific index series. These series were compared to each other, living-tree and subfossil tree-rings, and to early and modern meteorological time series. Phenological indices showed strong positive correlation with February to June temperatures. On the other hand, the correlations between phenological indices and precipitation data were around zero. Analysis using time-dependent running correlations showed non-stationary relationship between the tree-rings and phenological indices and observed spring temperatures. The skill of phenological data for reconstructing the spring temperatures was statistically proved.     

Bayesian Predictive Probability Functions for Count Data that are Subject to Misclassification

We develop three Bayesian predictive probability functions based on data in the form of a double sample. One Bayesian predictive probability function is for predicting the true unobservable count of interest in a future sample for a Poisson model with data subject to misclassification and two Bayesian predictive probability functions for predicting the number of misclassified counts in a current observable fallible count for an event of interest. We formulate a Gibbs sampler to calculate prediction intervals for these three unobservable random variables and apply our new predictive models to calculate prediction intervals for a real‐data example. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Aggression with mixed age-sex groups of bolivian squirrel monkeys following single animal introductions and new group formations

The social organization of the Bolivian squirrel monkey (Saimiri boliviensis boliviensis) is thought to be sexually segregated, with males and females forming separate social groups during the nonbreeding season. To investigate the influence of this social order on patterns of aggression, controlled single animal introductions within established groups and establishment of new groups were studied in a systematic manner. Behavioral observations were made using an all-occurrences technique to sample all agonistic interactions. In study I, 4 animals of each age-sex class (adult males and females, juvenile males and females) were introduced one at a time into 4 different social groups composed of 1 adult male and 8 to 10 adult females. Behavioral observations were made prior to and after the introductions. Newly introduced adult males received significantly more contact aggression than other age classes. Only the adult females increased aggression after the introduction of new animals. In study II, new social groups were formed and behavioral observations were made following formation. One group was formed from 7 pairs of familiar females and an unfamiliar adult male. This group had a high frequency of aggression during the first half hour, with contact aggression rising to peaks at 3 and 5 h. The second group consisted of 10 familiar females, 2 pairs of familiar females and an unfamiliar male. There was a significant peak in contact aggression 3 h into the observation.     

A foliar disease model for use in wheat disease management decision support systems

A model of winter wheat foliar disease is described, parameterised and tested for Septoria tritici (leaf blotch), Puccinia striiformis (yellow rust), Erysiphe graminis (powdery mildew) and Puccinia triticina (brown rust). The model estimates disease‐induced green area loss, and can be coupled with a wheat canopy model, in order to estimate remaining light‐intercepting green tissue and hence the capacity for resource capture. The model differs from those reported by other workers in three respects. First, variables (such as weather, host resistance and inoculum pressure) that affect disease risk are integrated in their effect on disease progress. The agronomic and meteorological data called for are restricted to those commonly available to growers by their own observations and from meteorological service networks. Second, field observations during the growing season can be used both to correct current estimates of disease severity and to modify parameters that determine predicted severity. Third, pathogen growth and symptom expression are modelled to allow the effects of fungicides to be accounted for as protectant activity (reducing infections that occur postapplication) and eradicant activity (reducing growth of presymptomatic infections). The model was tested against data from a wide range of sites and varieties and was shown to predict the expected level of disease sufficiently accurately to support fungicide treatment decisions.     

Selection of Judges for Sensory Evaluation

A test based on ordered observations for selection of consistent judges for sensory evaluation has been given. The test–statistic depends on the sum of ranks of different products under consideration. The probability distribution of the test-statistic has been worked out for small sample and it turns out to be chi-square distribution for large sample. The analytical procedure has been explained by a numerical example of a taste-testing experiment.     

Assessing Local Influence in Measurement Error Models

We consider the assessment of local influence for generalized linear models when the covariates are measured with errors. We show how to evaluate the effect that perturbations to the data, case weights, and model assumptions may have on the parameter estimates. Based on the likelihood displacement functions, some useful influence diagnostics are derived. Two examples illustrate application of the proposed diagnostics and assessment of the measurement error assumptions.     

A Mixed Model for Binary Data: The Modified X2-test

In this paper, a statistical model for clinical trials is presented for the special situation that a varying and unstructered number of binary responses is obtained from each subject. The assumptions of the model are the following: 1.) For each subject there is a (constant) individual Bernoulli parameter determining the distribution of the binary responses of this subject. 2.) The Bernoulli parameters associated with the subjects are realizations of independent random variables with distributions P_g in treatment group g(g = 1, 2, …, G). 3.) Given the value of the Bernoulli parameter, the observations are stochastically independent within each subject. Under these assumptions, a test statistic is derived to test the hypothesis H₀:E(P₁) = E(P₂) = … = E(P_G). It is proven and demonstrated by simulations, that the test statistic asymptotically (i.e. for a large number of subjects) follows the X²-distribution.     

Combining palaeolimnological and limnological approaches in assessing lake ecosystem response to nutrient reduction

1. Palaeolimnological data and limnological time‐series data are highly complementary. Sediment records extend time‐scales, integrate subannual variability and expand the range of sites that can be studied, but they suffer from taphonomic biases and occasionally from uncertain chronology. Observational time‐series data, on the other hand, are highly resolved but are very limited in extent both in space and time. 2. Palaeolimnological and observational data‐sets need to be combined in oligotrophication research to establish (i) the past and present status of lakes needed to identify reference conditions; (ii) changes in ecosystem state; (iii) responses to nutrient reduction; and (iv) the potential role of other factors (e.g. additional stressors, climate change) that might confound predictions of future state.     

洛浦县山普拉出土颅骨的初步研究

本文研究洛浦县山普拉公社出土的成人颅骨59例(男26、女33),其时代为公元前一世纪—公元三世纪。山普拉人显示出既有大蒙古人种的大部分特征,也有欧罗巴人种的许多特征。山普拉人是一个混血的民族。根据发掘情况,可以肯定他们是当地土著。经t值检验有半数测量性体征在两性之间有非常显著或显著差异。     

On the COLTON Model for Clinical Trials with Delayed Observations-Dichotomous Responses

In the COLTON model for clinical trials a choice is to be made between two medical treatments where there is a known patient horizon N. In an earlier paper we studied the COLTON model under the additional assumption that there is a time lag between the administration of the treatments and the availability of the responses where the responses are normally distributed. Here we extend the results of the earlier paper to the case where responses are dichotomous. The relative performance of two simple procedures for dealing with the patients who arrive during the waiting period between the end of the administration of treatment in the trial phase and observation of the final trial response is discussed within a BAYESIAN framework.     

A Nonparametric Test for Ordered Alternatives in Two-Way Layouts with Censored Data

A nonparametric test is presented to test for ordered alternatives of the treatment effects when the data follow a two-way layout. The observations are subject to arbitrary right censorship. The test is similar to the one developed by Hettmansperger and Norton (1987) for uncensored observations.     

Estimation of Several Missing Values in F-Square Designs

The present study deals with the estimation of several missing values in F-square designs. The estimating equations for the non-iterative least squares estimation of Missing Values and explicit expressions for the estimators of the particular patterns of Missing Values are presented. This procedure is illustrated with the help of a numerical example.     

Spatial and temporal habitat use and selection by red deer: The use of direct and indirect methods

Given the importance of red deer Cervus elaphus for hunting and conservation purposes, understanding the interactions between this species and its habitats in the Mediterranean region is a crucial step for the sustainable management of this species. Aiming to compare pellet group counts and direct observations methods to study the habitat use and selection by red deer, the results obtained by both methods were compared, and their advantages and disadvantages were discussed. To understand the temporal patterns of habitat use and selection, the survey was conducted at three different seasons, birth period, rut season and winter. The habitat use and selection were studied in relation to land cover, watercourse, roads, ecotone zones and other topographic features (altitude, slope and aspect), using generalized linear models and selection ratios. The similarity of the results provided by pellet group counts and direct observations indicate that both methods may constitute useful tools to study the habitat use and selection by red deer. Globally, red deer seemed to select habitats that provide simultaneously food and some cover, as shown by its preference for shrublands, independently of the sampling season. The positive selection of ecotone zones embodies the need for open spaces. Males and females showed a similar use of shrubland, but selected patches with different characteristics therein. The spatial and temporal patterns exhibited by our results suggest that red deer balance their habitat requirements in respect to each phase of their reproductive cycle. Pellet group counts and direct observations seem to be useful methods to analyze habitat use and selection, and may provide helpful knowledge to the management and conservation of red deer.     

龋坏引起的慢性牙髓炎牙髓组织中的细菌超微观察

对１０例龋坏引起的慢性牙髓炎、未穿髓的新鲜离体患牙,进行分段定位、半薄切片、革兰氏染色,光镜观察定位后,再行超薄切片,醋酸铀－柠檬酸铅复染,电镜进行超微结构观察。结果：（１）患牙牙髓组织中有５例观察到多种类型微生物存在,为球杆菌、球菌、杆菌。（２）牙髓组织中的细菌位于细胞内及细胞外,并处于各种状态,有生长繁殖阶段的也有固缩溶解阶段的。（３）深龋引起的长期有症状的慢性牙髓炎牙髓组织中有大量活细菌存在,细菌均存在于冠髓内,接近龋坏部位,周围及深层组织中未见菌体。并观察到组织细胞吞噬大量细菌后,细胞变性,出现核固缩、核仁消失,细胞膜不清楚。细胞周围有大量溶酶体出现。结果提示：深龋引起的长期有症状的慢性牙髓炎,尽管未穿髓,牙髓组织中,仍可有大量活细菌长期存在。     

CoelacanthLatimeria chalumnae aggregates in caves: first observations on their resting habitat and social behavior

Synopsis We report first observations on social behavior ofLatimeria chalumnae in its natural environment at around 200 m depth in the Comoro Archipelago, Western Indian Ocean. Coelacanths aggregate in small nonaggressive groups in sheltered lava-caves. They live in open site-attached social systems with individuals occupying large overlapping home ranges of more than 8 km coastline, some for a period of at least 2 years.     

Analysis of a partially observed binary covariate process and a censored failure time in the presence of truncation and competing risks

We develop methods for assessing the association between a binary time-dependent covariate process and a failure time endpoint when the former is observed only at a single time point and the latter is right censored, and when the observations are subject to truncation and competing causes of failure. Using a proportional hazards model for the effect of the covariate process on the failure time of interest, we develop an approach utilizing EM algorithm and profile likelihood for estimating the relative risk parameter and cause-specific hazards for failure. The methods are extended to account for other covariates that can influence the time-dependent covariate process and cause-specific risks of failure. We illustrate the methods with data from a recent study on the association between loss of hepatitis B e antigen and the development of hepatocellular carcinoma in a population of chronic carriers of hepatitis B.     

A generalized Friedman test for data defined by intervals

If observations are restricted in such a way that only intervals covering the true but hidden values are known—situations found in clinical studies or analyzing laboratory data-standard nonpara-metric procedures fail. The present paper suggests the use of generalized Friedman tests, if the sample consists of nk observations, with one observation from each of k treatments in each of in blocks. Furthermore the proposed test enables Friedman-type analyses of censored data. An example is given and a computer program facilitating the use of the test is introduced.     

Note on combining risks using a logistic discriminant function approach

A simple method for combining risks based on repeated observations on an individual is presented. Implications of this model are explored for several special cases, and an example is presented.