Finite sample nonparametric tests for linear regressions

We introduce several exact nonparametric tests for finite sample multivariatelinear regressions, and compare their powers. This fills an important gap inthe literature where the only known nonparametric tests are either asymptotic,or assume one covariate only.

In vitro assessment of the pulmonary toxicity and gastric availability of lead-rich particles from a lead recycling plant

Epidemiological studies in urban areas have linked increasing respiratory and cardiovascular pathologies with atmospheric particulate matter (PM) from anthropic activities. However, the biological fate of metal-rich PM industrial emissions in urban areas of developed countries remains understudied. Lead toxicity and bioaccessibility assessments were therefore performed on emissions from a lead recycling plant, using complementary chemical acellular tests and toxicological assays, as a function of PM size (PM(10-2.5), PM(2.5-1) and PM(1)) and origin (furnace, refining and channeled emissions). Process PM displayed differences in metal content, granulometry, and percentage of inhalable fraction as a function of their origin. Lead gastric bioaccessibility was relatively low (maximum 25%) versus previous studies; although, because of high total lead concentrations, significant metal quantities were solubilized in simulated gastrointestinal fluids. Regardless of origin, the finest PM(1) particles induced the most significant pro-inflammatory response in human bronchial epithelial cells. Moreover, this biological response correlated with pro-oxidant potential assay results, suggesting some biological predictive value for acellular tests. Pulmonary effects from lead-rich PM could be driven by thiol complexation with either lead ions or directly on the particulate surface. Finally, health concern of PM was discussed on the basis of pro-inflammatory effects, accellular test results, and PM size distribution.

Multi-sample analysis of moment-structures: Asymptotic validity of inferences based on second-order moments

In moment structure analysis with nonnormal data, asymptotic valid inferences require the computation of a consistent (under general distributional assumptions) estimate of the matrix $\Gamma$ of asymptotic variances of sample second--order moments. Such a consistent estimate involves the fourth--order sample moments of the data. In practice, the use of fourth--order moments leads to computational burden and lack of robustness against small samples. In this paper we show that, under certain assumptions, correct asymptotic inferences can be attained when $\Gamma$ is replaced by a matrix $\Omega$ that involves only the second--order moments of the data. The present paper extends to the context of multi--sample analysis of second--order moment structures, results derived in the context of (simple--sample) covariance structure analysis (Satorra and Bentler, 1990). The results apply to a variety of estimation methods and general type of statistics. An example involving a test of equality of means under covariance restrictions illustrates theoretical aspects of the paper.

Impact of biocontrol strain Pseudomonas fluorescens CHA0 on rhizosphere bacteria isolated from barley (Hordeum vulgare L.) with special reference to Cytophaga-like bacteria.

AIMS: To assess the impact of the biocontrol strain Pseudomonas fluorescens CHA0 on a collection of barley rhizosphere bacteria using an agar plate inhibition assay and a plant microcosm, focusing on a CHA0-sensitive member of the Cytophaga-like bacteria (CLB). METHODS AND RESULTS: The effect of strain CHA0 on a collection of barley rhizosphere bacteria, in particular CLB and fluorescent pseudomonads sampled during a growth season, was assessed by a growth inhibition assay. On average, 85% of the bacteria were sensitive in the May sample, while the effect was reduced to around 68% in the July and August samples. In the May sample, around 95% of the CLB and around 45% of the fluorescent pseudomonads were sensitive to strain CHA0. The proportion of CHA0-sensitive CLB and fluorescent pseudomonad isolates decreased during the plant growth season, i.e. in the July and August samples. A particularly sensitive CLB isolate, CLB23, was selected, exposed to strain CHA0 (wild type) and its genetically modified derivatives in the rhizosphere of barley grown in gnotobiotic soil microcosms. Two dry-stress periods were imposed during the experiment. Derivatives of strain CHA0 included antibiotic or exopolysaccharide (EPS) overproducing strains and a dry-stress-sensitive mutant. Despite their inhibitory activity against CLB23 in vitro, neither wild-type strain CHA0, nor any of its derivatives, had a major effect on culturable and total cell numbers of CLB23 during the 23-day microcosm experiment. Populations of all inoculants declined during the two dry-stress periods, with soil water contents below 5% and plants reaching the wilting point, but they recovered after re-wetting the soil. Survival of the dry-stress-sensitive mutant of CHA0 was most affected by the dry periods; however, this did not result in an increased population density of CLB23. CONCLUSIONS: CLB comprise a large fraction of barley rhizosphere bacteria that are sensitive to the biocontrol pseudomonad CHA0 in vitro. However, in plant microcosm experiments with varying soil humidity conditions, CHA0 or its derivatives had no major impact on the survival of the highly sensitive CLB strain, CLB23, during two dry-stress periods and a re-wetting period; all co-existed well in the rhizosphere of barley plants. SIGNIFICANCE AND IMPACT OF THE STUDY: Results indicate a lack of interaction between the biocontrol pseudomonad CHA0 and a sensitive CLB when the complexity increases from agar plate assays to plant microcosm experiments. This suggests the occurrence of low levels of antibiotic production and/or that the two bacterial genera occupy different niches in the rhizosphere.

Scaled and adjusted restricted tests in multi-sample analysis of moment structures

We extend to score, Wald and difference test statistics the scaled and adjusted corrections to goodness-of-fit test statistics developed in Satorra and Bentler (1988a,b). The theory is framed in the general context of multisample analysis of moment structures, under general conditions on the distribution of observable variables. Computational issues, as well as the relation of the scaled and corrected statistics to the asymptotic robust ones, is discussed. A Monte Carlo study illustrates thecomparative performance in finite samples of corrected score test statistics.

Bringing game theory to hypothesis testing: Establishing finite sample bounds on inference

Small sample properties are of fundamental interest when only limited data is avail-able. Exact inference is limited by constraints imposed by speci.c nonrandomizedtests and of course also by lack of more data. These e¤ects can be separated as we propose to evaluate a test by comparing its type II error to the minimal type II error among all tests for the given sample. Game theory is used to establish this minimal type II error, the associated randomized test is characterized as part of a Nash equilibrium of a .ctitious game against nature.We use this method to investigate sequential tests for the di¤erence between twomeans when outcomes are constrained to belong to a given bounded set. Tests ofinequality and of noninferiority are included. We .nd that inference in terms oftype II error based on a balanced sample cannot be improved by sequential sampling or even by observing counter factual evidence providing there is a reasonable gap between the hypotheses.

Some hypothesis tests for the covariance matrix when the dimension is large compared to the sample size

This paper analyzes whether standard covariance matrix tests work whendimensionality is large, and in particular larger than sample size. Inthe latter case, the singularity of the sample covariance matrix makeslikelihood ratio tests degenerate, but other tests based on quadraticforms of sample covariance matrix eigenvalues remain well-defined. Westudy the consistency property and limiting distribution of these testsas dimensionality and sample size go to infinity together, with theirratio converging to a finite non-zero limit. We find that the existingtest for sphericity is robust against high dimensionality, but not thetest for equality of the covariance matrix to a given matrix. For thelatter test, we develop a new correction to the existing test statisticthat makes it robust against high dimensionality.

Honey, I shrunk the sample covariance matrix

The central message of this paper is that nobody should be using the samplecovariance matrix for the purpose of portfolio optimization. It containsestimation error of the kind most likely to perturb a mean-varianceoptimizer. In its place, we suggest using the matrix obtained from thesample covariance matrix through a transformation called shrinkage. Thistends to pull the most extreme coefficients towards more central values,thereby systematically reducing estimation error where it matters most.Statistically, the challenge is to know the optimal shrinkage intensity,and we give the formula for that. Without changing any other step in theportfolio optimization process, we show on actual stock market data thatshrinkage reduces tracking error relative to a benchmark index, andsubstantially increases the realized information ratio of the activeportfolio manager.

Impact of a gall midge Parkiamyia paraensis (Diptera, Cecidomyiidae) on the Amazonian plant Parkia pendula (Fabaceae)

We observed the occurrence of large numbers of galls induced by Parkiamyia paraensis (Diptera, Cecidomyiidae) on the leaflets of Parkia pendula (Fabaceae) in northern Para, Brazil. We addressed two questions in this study: i) what is the proportion of attacked plants in the field, and nursery conditions?; and ii) what is the impact of galls on the host plant? An average of 86% of the plants were galled in the field. Galled P. pendula were distinct from healthy individuals due to their prostrated architecture and death of terminal shoots. Approximately 50% of the total available leaves and 35% leaflets were attacked by P. paraensis on saplings under nursery conditions. Each one-year old plant supported an average of 1,300 galls, and an average of 60g allocated to galled tissue. Otherwise, attacked individuals were taller and heavier than healthy plants. Attacked plants weighed five times more than healthy plants. When the weight of the galls was removed, the total weight (aerial part without galls) of attacked plants was drastically reduced, indicating that most of the biomass of attacked plants was due to the attack by P. paraensis galls. Although the data indicate a paradox, as young plants attacked by the galling herbivore appear to develop more vigorously than unattacked plants, we suggest that P. paraensis negatively affect P. pendula development.

Evolution in subdivided plant populations: concepts, recent advances and future directions.

SUMMARY: Research into the evolution of subdivided plant populations has long involved the study of phenotypic variation across plant geographic ranges and the genetic details underlying that variation. Genetic polymorphism at different marker loci has also allowed us to infer the long- and short-term histories of gene flow within and among populations, including range expansions and colonization-extinction dynamics. However, the advent of affordable genome-wide sequences for large numbers of individuals is opening up new possibilities for the study of subdivided populations. In this review, we consider what the new tools and technologies may allow us to do. In particular, we encourage researchers to look beyond the description of variation and to use genomic tools to address new hypotheses, or old ones afresh. Because subdivided plant populations are complex structures, we caution researchers away from adopting simplistic interpretations of their data, and to consider the patterns they observe in terms of the population genetic processes that have given rise to them; here, the genealogical framework of the coalescent will continue to be conceptually and analytically useful.

Relationships between host plant architecture and gall abundance and survival

The plant architecture hypothesis predicts that variation in host plant architecture influences insect herbivore community structure, dynamics and performance. In this study we evaluated the effects of Macairea radula (Melastomataceae) architecture on the abundance of galls induced by a moth (Lepidoptera: Gelechiidae). Plant architecture and gall abundance were directly recorded on 58 arbitrarily chosen M. radula host plants in the rainy season of 2006 in an area of Cerrado vegetation, southeastern Brazil. Plant height, dry biomass, number of branches, number of shoots and leaf abundance were used as predicting variables of gall abundance and larval survival. Gall abundance correlated positively with host plant biomass and branch number. Otherwise, no correlation (p > 0.05) was found between gall abundance with shoot number or with the number of leaves/plant. From a total of 124 galls analyzed, 67.7% survived, 14.5% were attacked by parasitoids, while 17.7% died due to unknown causes. Larvae that survived or were parasitized were not influenced by architectural complexity of the host plant. Our results partially corroborate the plant architecture hypothesis, but since parasitism was not related to plant architecture it is argued that bottom-up effects may be more important than top-down effects in controlling the population dynamics of the galling lepidopteran. Because galling insects often decrease plant fitness, the potential of galling insects in selecting for less architectural complex plants is discussed.