Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “statistical inference”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,333 records · Page 74Linked to original sources
PubMed2001

The exploitation of regularities in the environment by the brain.

Statistical regularities of the environment are important for learning, memory, intelligence, inductive inference, and in fact, for any area of cognitive science where an information-processing brain promotes survival by exploiting them. This has been recognised by many of those interested in cognitive function, starting with Helmholtz, Mach, and Pearson, and continuing through Craik, Tolman, Attneave, and Brunswik. In the current era, many of us have begun to show how neural mechanisms exploit the regular statistical properties of natural images. Shepard proposed that the apparent trajectory of an object when seen successively at two positions results from internalising the rules of kinematic geometry, and although kinematic geometry is not statistical in nature, this is clearly a related idea. Here it is argued that Shepard's term, "internalisation," is insufficient because it is also necessary to derive an advantage from the process. Having mechanisms selectively sensitive to the spatio-temporal patterns of excitation commonly experienced when viewing moving objects would facilitate the detection, interpolation, and extrapolation of such motions, and might explain the twisting motions that are experienced. Although Shepard's explanation in terms of Chasles' rule seems doubtful, his theory and experiments illustrate that local twisting motions are needed for the analysis of moving objects and provoke thoughts about how they might be detected.

Brain↗
PubMed2005

Dynamic Trees for unsupervised segmentation and matching of image regions.

We present a probabilistic framework--namely, multiscale generative models known as Dynamic Trees (DT)--for unsupervised image segmentation and subsequent matching of segmented regions in a given set of images. Beyond these novel applications of DTs, we propose important additions for this modeling paradigm. First, we introduce a novel DT architecture, where multilayered observable data are incorporated at all scales of the model. Second, we derive a novel probabilistic inference algorithm for DTs--Structured Variational Approximation (SVA)--which explicitly accounts for the statistical dependence of node positions and model structure in the approximate posterior distribution, thereby relaxing poorly justified independence assumptions in previous work. Finally, we propose a similarity measure for matching dynamic-tree models, representing segmented image regions, across images. Our results for several data sets show that DTs are capable of capturing important component-subcomponent relationships among objects and their parts, and that DTs perform well in segmenting images into plausible pixel clusters. We demonstrate the significantly improved properties of the SVA algorithm--both in terms of substantially faster convergence rates and larger approximate posteriors for the inferred models--when compared with competing inference algorithms. Furthermore, results on unsupervised object recognition demonstrate the viability of the proposed similarity measure for matching dynamic-structure statistical models.

Algorithms↗
PubMed1992

Methodologic issues in using epidemiologic studies of occupational cohorts for cancer risk assessment.

This paper focuses on presenting a review and discussion of the major methodologic issues involved in using epidemiologic studies of occupational groups for assessing human cancer risks. Although animal studies have been most often used for quantitative risk assessment, it is generally recognized that well conducted epidemiologic studies would provide the best basis for estimating human risk. However, there are several features related to the design and analysis of epidemiologic studies which frequently limit their usefulness for quantitating risks. The lack of accurate information on exposure in epidemiologic studies is perhaps the most frequently cited limitation of these studies for risk assessment. However, other features of epidemiologic study design such as statistical power, length of follow-up, selection bias, confounding and effect modification may also limit the inferences that can be drawn from these studies. Furthermore even when the aforementioned limitations are overcome, substantial uncertainty exists concerning the choice of an appropriate statistical (or biologic) model for extrapolation beyond the range of exposures observed in a particular study. An empirical example is provided in which estimates of risk varied by nearly 3 orders of magnitude depending on which functional form of the model was chosen. Modeling of epidemiologic data for QRA should be based upon internal comparisons rather than on modeling Standardized Mortality Ratios (SMRs) when possible. Because of the limitations discussed in this paper, epidemiologic data should not be viewed as a panacea for the problems inherent in using animal bioassay data for QRA.(ABSTRACT TRUNCATED AT 250 WORDS)

Bias↗
PubMed2004

Linear models and empirical bayes methods for assessing differential expression in microarray experiments.

The problem of identifying differentially expressed genes in designed microarray experiments is considered. Lonnstedt and Speed (2002) derived an expression for the posterior odds of differential expression in a replicated two-color experiment using a simple hierarchical parametric model. The purpose of this paper is to develop the hierarchical model of Lonnstedt and Speed (2002) into a practical approach for general microarray experiments with arbitrary numbers of treatments and RNA samples. The model is reset in the context of general linear models with arbitrary coefficients and contrasts of interest. The approach applies equally well to both single channel and two color microarray experiments. Consistent, closed form estimators are derived for the hyperparameters in the model. The estimators proposed have robust behavior even for small numbers of arrays and allow for incomplete data arising from spot filtering or spot quality weights. The posterior odds statistic is reformulated in terms of a moderated t-statistic in which posterior residual standard deviations are used in place of ordinary standard deviations. The empirical Bayes approach is equivalent to shrinkage of the estimated sample variances towards a pooled estimate, resulting in far more stable inference when the number of arrays is small. The use of moderated t-statistics has the advantage over the posterior odds that the number of hyperparameters which need to estimated is reduced; in particular, knowledge of the non-null prior for the fold changes are not required. The moderated t-statistic is shown to follow a t-distribution with augmented degrees of freedom. The moderated t inferential approach extends to accommodate tests of composite null hypotheses through the use of moderated F-statistics. The performance of the methods is demonstrated in a simulation study. Results are presented for two publicly available data sets.

Journal Article↗
PubMed1997

Mind your manners. Part III: Individual scenario results and discussion of the National Association of Medical Examiners Manner of Death Questionnaire, 1995.

In 1995, a questionnaire was distributed to the > 700 physician medical examiner/coroners (ME/Cs) who are members of the National Association of Medical Examiners (NAME, St. Louis, MO, U.S.A.). The questionnaire consisted of 23 death scenarios for which individual responders were asked to assign a manner of death (homicide, suicide, accident, natural, or undetermined); 198 questionnaires were completed and analyzed. The distribution of manner of death responses was tabulated. In addition, a nosologist from the National Center for Health Statistics was provided with a cause-of-death statement based on each scenario and was asked to assign an International Classification of Diseases (ICD) code for the underlying cause of death, from which a manner of death was inferred from the ICD code's literal text description. Overall, agreement on a given manner of death in a single scenario was > 90% in only 4 of 23 scenarios and > 70% in only 12 of 23 scenarios. However, in 21 scenarios, the most common response comprised a majority. The manner of death inferred from the ICD code that was assigned by the National Center for Health Statistics (NCHS) matched the most common response of participants in 18 of the 23 scenarios. The questionnaire results show that there is substantial disagreement among experienced MEs concerning the manner of death classification that is preferred for selected types of death. Encouraging, however, is the fact that the manner of death coded for statistical purposes generally agreed with the most common classification of manner made by ME/Cs. Highlights from the discussion of each scenario that occurred during the NAME interim meeting (Nashville, Tennessee, February 1996) are also included. Other portions of the program including history of manner of death concepts and results of questions regarding responder training and characteristics are published separately in this issue of the Journal. Information derived from the questionnaire should be useful to those planning strategies to improve the consistency of manner of death classifications by ME/Cs.

Autopsy↗
PubMed1990

An assessment of the impact of ambient ozone on field-grown crops in New Jersey using the EDU method: part 2-soybean (Glycine max (L.) Merr.).

The yields of eleven commercially grown soybean cultivars were compared in ethylenediurea (EDU)-treated and non-treated field plots in New Brunswick, New Jersey, over a 4 year period. No statistically significant difference between treatments was found for any cultivar; the inference being ambient ozone did not adversely affect soybean yield. Succeeding field experiments supported this interpretation of the data. 'Sanilac' white bean, a legume known to be more sensitive to O(3) than soybean, was found to produce a significantly greater yield in EDU-treated than non-treated plots, unlike a companion planting of 'Williams 82' soybean which did not exhibit the differential response. The results indicated that the specific EDU protocol used in the soybean experiments is capable of detecting an ozone effect in a legume. Moreover, in a concurrent greenhouse experiment the yield of EDU-treated Sanilac white bean was not significantly different from non-treated plants in the absence of ozone pollution. In a dose-response field experiment during a year of unusually high O(3) pollution, yield of 'Williams 82' increased slightly with each EDU increment up to 500 ppm and decreased at 1000 ppm. The difference between non-treated and EDU-treated plants, however, was not statistically significant. There was no evidence to suggest that the EDU concentration (500 ppm) used in previous soybean experiments reduced seed yield. Fortuitously, the tolerance of commercially-grown soybean to ambient ozone is at least partially conditioned by the practce of not irrigating the crop. The New Jersey results are in agreement with reports from Maryland, Georgia and Tennessee in which an adverse impact of ambient O(3) was not found in soybean, but contrary to a current predictive model.

Journal Article↗
PubMedDate not supplied

Some statistical issues in modelling pharmacokinetic data.

A fundamental assumption underlying pharmacokinetic compartment modelling is that each subject has a different individual curve. To some extent this runs counter to the statistical principle that similar individuals will have similar curves, thus making inferences to a wider population possible. In population pharmacokinetics, the compromise is to use random effects. We recommend that such models also be used in data rich situations instead of independently fitting individual curves. However, the additional information available in such studies shows that random effects are often not sufficient; generally, an autoregressive process is also required. This has the added advantage that it provides a means of tracking each individual, yielding predictions for the next observation. The compartment model curve being fitted may also be distorted in other ways. A widely held assumption is that most, if not all, pharmacokinetic concentration data follow a log-normal distribution. By examples, we show that this is not generally true, with the gamma distribution often being more suitable. When extreme individuals are present, a heavy-tailed distribution, such as the log Cauchy, can often provide more robust results. Finally, other assumptions that can distort the results include a direct dependence of the variance, or other dispersion parameter, on the mean and setting non-detectable values to some arbitrarily small value instead of treating them as censored. By pointing out these problems with standard methods of statistical modelling of pharmacokinetic data, we hope that commercial software will soon make more flexible and suitable models available.

Humans↗
PubMed2004

Tempo of speciation in a butterfly genus from the Southeast Asian tropics, inferred from mitochondrial and nuclear DNA sequence data.

Molecular systematics is frequently beset with phylogenetic results that are not fully resolved. Researchers either state that the absence of resolution is due to character conflict, explosive speciation, or some combination of the two, but seldom do they carefully examine their data to distinguish between these causes. In this study, we exhaustively analyze a set of nuclear and mitochondrial nucleotide data for the Asian tropical butterfly genus Arhopala so as to highlight the causes of polytomies in the phylogenetic trees, and, as a result, to infer important biological events in the history of this genus. We began by using non-parametric statistical methods to determine whether the ambiguously resolved regions in these trees represent hard or soft polytomies. In addition we determined how this correlated to number of inferred changes on branches, using parametric maximum likelihood estimations. Based on congruent patterns in both mitochondrial and nuclear DNA sequences, we concluded that at two stages in the history of Arhopala there have been accelerated instances of speciation. One event, at the base of the phylogeny, generated many of the groups and subgroups currently recognized in this genus, while a later event generated another major clade consisting of both Oriental and Papuan species groups. Based on comparisons of closely related taxa, the ratio of instantaneous rate of evolution between mitochondrial and nuclear DNA evolution is established at approximately 3:1. The earliest radiation is dated between 7 and 11 Ma by a molecular clock analysis, setting the events generating much of the diversity of Arhopala at well before the Pleistocene. Periodical flooding of the Sunda plateau during interglacial periods was, therefore, not responsible for generating the major divisions in the genus Arhopala. Instead, we hypothesize that large-scale climatic changes taking place in the Miocene have induced the early acceleration in speciation.

Animals↗
PubMed1994

Beware of using small statistical samples when assessing the quality of a DNA library.

DNA libraries often contain very large numbers of clones (from 1000 up to 700,000). Since at present it is impossible to analyze all of these clones, usually statistical samples comprising less than 100 clones are tested. The quality of the library is then assessed by linear extrapolation. Occasionally, full coverage of chromosomal regions by DNA probes is inferred from this. However, this may not be accurate since linear extrapolation is misleading and the statistical samples are generally too small to characterize the libraries. A quantitative model of the distribution of the frequencies of the clones in a library is mandatory for any useful assessment of the quality of the library. Otherwise, it is very difficult to draw useful conclusions from moderately sized samples. Examples from everyday life and formulas are given to determine the quality of a library and useful sample sizes.

Chromosome Walking↗
PubMed1979

Statistical errors in papers in the British Journal of Psychiatry.

In the 12 months from July 1977 to June 1978, out of 168 papers in the British Journal of Psychiatry, 139 (83 per cent) presented numerical results. A total of 63 papers contained statistical errors and at least one drew unsupportable conclusions. In many cases the errors were not considered to be severe, but they were often sufficient to raise doubts about some inferences. It is suggested that more consultation of investigators with medical statisticians, a basic training in statistical methods in the preclinical years, and more extensive statistical refereeing are needed.

Periodicals as Topic↗
PubMed2000

High prevalence of Helicobacter heilmannii-associated gastritis in a small, predominantly rural area: further evidence in support of a zoonosis?

BACKGROUND: Primary hosts of Helicobacter heilmannii are domestic animals--cats, dogs and pigs, but rarely is it detected in gastric biopsies from humans. We found H. heilmannii in gastric biopsies obtained from patients living in a predominantly rural area. METHODS: We evaluated geographic and demographic data from the area and calculated both the total prevalence and the prevalence in each community in this area. Chi-squared test and exploratory data analysis were used for statistical evaluation. Histologic and clinical data were recorded. RESULTS: Forty-three communities, mostly rural, were identified in the area, the size of which is about 200 km2. H. heilmannii was detected in 33 patients (prevalence 2%) living in 20 communities. The prevalence of H. heilmannii statistically significantly varied in those communities from 0.06%-1.1%. It was possible to infer that there is a negative correlation between prevalence and community size. Chronic active gastritis was diagnosed in all patients. The active inflammation became inactive after eradication of H. heilmannii via common anti-Helicobacter therapy. CONCLUSION: A high prevalence of H. heilmannii-associated gastritis in a small, predominantly rural area and statistically significant differences in prevalence from one community to the next, i.e. highest in the smallest village and lowest in a small town with a rather urban lifestyle, were found. In our opinion, these prevalence rates may be a consequence of transmission of the infection from domestic animals, because animal-human contact is generally more common in villages than in towns.

Animals↗
PubMed2005

Modularized learning of genetic interaction networks from biological annotations and mRNA expression data.

MOTIVATION: Inferring the genetic interaction mechanism using Bayesian networks has recently drawn increasing attention due to its well-established theoretical foundation and statistical robustness. However, the relative insufficiency of experiments with respect to the number of genes leads to many false positive inferences. RESULTS: We propose a novel method to infer genetic networks by alleviating the shortage of available mRNA expression data with prior knowledge. We call the proposed method 'modularized network learning' (MONET). Firstly, the proposed method divides a whole gene set to overlapped modules considering biological annotations and expression data together. Secondly, it infers a Bayesian network for each module, and integrates the learned subnetworks to a global network. An algorithm that measures a similarity between genes based on hierarchy, specificity and multiplicity of biological annotations is presented. The proposed method draws a global picture of inter-module relationships as well as a detailed look of intra-module interactions. We applied the proposed method to analyze Saccharomyces cerevisiae stress data, and found several hypotheses to suggest putative functions of unclassified genes. We also compared the proposed method with a whole-set-based approach and two expression-based clustering approaches.

Algorithms↗
PubMed2000

Saddlepoint approximations for small sample logistic regression problems.

Double saddlepoint approximations provide quick and accurate approximations to exact conditional tail probabilities in a variety of situations. This paper describes the use of these approximations in two logistic regression problems. An investigation of regression analysis of the log-odds ratio in a sequence or set of 2x2 tables via simulation studies shows that in practical settings the saddlepoint methods closely approximate exact conditional inference. The double saddlepoint approximation in the test for trend in a sequence of binomial random variates is also shown, via simulation studies, to be an effective approximation to exact conditional inference.

Logistic Models↗
PubMed2004

A spline function approach for detecting differentially expressed genes in microarray data analysis.

MOTIVATION: A primary objective of microarray studies is to determine genes which are differentially expressed under various conditions. Parametric tests, such as two-sample t-tests, may be used to identify differentially expressed genes, but they require some assumptions that are not realistic for many practical problems. Non-parametric tests, such as empirical Bayes methods and mixture normal approaches, have been proposed, but the inferences are complicated and the tests may not have as much power as parametric models. RESULTS: We propose a weakly parametric method to model the distributions of summary statistics that are used to detect differentially expressed genes. Standard maximum likelihood methods can be employed to make inferences. For illustration purposes the proposed method is applied to the leukemia data (training part) discussed elsewhere. A simulation study is conducted to evaluate the performance of the proposed method.

Algorithms↗
PubMed2005

Differential gene expression detection and sample classification using penalized linear regression models.

Differential gene expression detection and sample classification using microarray data have received much research interest recently. Owing to the large number of genes p and small number of samples n (p >> n), microarray data analysis poses big challenges for statistical analysis. An obvious problem owing to the 'large p small n' is over-fitting. Just by chance, we are likely to find some non-differentially expressed genes that can classify the samples very well. The idea of shrinkage is to regularize the model parameters to reduce the effects of noise and produce reliable inferences. Shrinkage has been successfully applied in the microarray data analysis. The SAM statistics proposed by Tusher et al. and the 'nearest shrunken centroid' proposed by Tibshirani et al. are ad hoc shrinkage methods. Both methods are simple, intuitive and prove to be useful in empirical studies. Recently Wu proposed the penalized t/F-statistics with shrinkage by formally using the (1) penalized linear regression models for two-class microarray data, showing good performance. In this paper we systematically discussed the use of penalized regression models for analyzing microarray data. We generalize the two-class penalized t/F-statistics proposed by Wu to multi-class microarray data. We formally derive the ad hoc shrunken centroid used by Tibshirani et al. using the (1) penalized regression models. And we show that the penalized linear regression models provide a rigorous and unified statistical framework for sample classification and differential gene expression detection.

Algorithms↗
PubMed2003

Comparing hierarchical modeling with traditional logistic regression analysis among patients hospitalized with acute myocardial infarction: should we be analyzing cardiovascular outcomes data differently?

BACKGROUND: Data in health research are frequently structured hierarchically. For example, data may consist of patients treated by physicians who in turn practice in hospitals. Traditional statistical techniques ignore the possible correlation of outcomes within a given practice or hospital. Furthermore, imputing characteristics measured at higher levels of the hierarchy to the patient-level artificially inflates the amount of available information on the effect of higher-level characteristics on outcomes. METHODS: Conventional logistic regression models and multilevel logistic regression models were fit to a cross-sectional cohort of patients hospitalized with a diagnosis of acute myocardial infarction. The statistical significance of the effect of patient, physician, and hospital characteristics on patient outcomes was compared between the 2 modeling strategies. RESULTS: The 2 analytic strategies agreed well on the effect of patient characteristics on outcomes. According to the traditional analysis, teaching status was statistically significantly associated with 5 of the 9 outcomes, whereas the multilevel models did not find a statistically significant association between teaching status and any patient outcomes. Similarly, the traditional and multilevel models disagreed on the statistical significance of the effect of being treated at a revascularization hospital and 3 patient outcomes. CONCLUSIONS: In comparing the resultant models, we see that false inferences can be drawn by ignoring the structure of the data. Conventional logistic regression tended to increase the statistical significance for the effects of variables measured at the hospital-level compared to the level of significance indicated by the multilevel model.

Adult↗
PubMed2004

The burgeoning field of statistical phylogeography.

In the newly emerging field of statistical phylogeography, consideration of the stochastic nature of genetic processes and explicit reference to theoretical expectations under various models has dramatically transformed how historical processes are studied. Rather than being restricted to ad hoc explanations for observed patterns of genetic variation, assessments about the underlying evolutionary processes are now based on statistical tests of various hypotheses, as well as estimates of the parameters specified by the models. A wide range of demographical and biogeographical processes can be accommodated by these new analytical approaches, providing biologically more realistic models. Because of these advances, statistical phylogeography can provide unprecedented insights about a species' history, including decisive information about the factors that shape patterns of genetic variation, species distributions, and speciation. However, to improve our understanding of such processes, a critical examination and appreciation of the inherent difficulties of historical inference and challenges specific to testing phylogeographical hypotheses are essential. As the field of statistical phylogeography continues to take shape many difficulties have been resolved. Nonetheless, careful attention to the complexities of testing historical hypotheses and further theoretical developments are essential to improving the accuracy of our conclusions about a species' history.

Demography↗
PubMed2004

Mapping quantitative trait loci with censored observations.

The existing statistical methods for mapping quantitative trait loci (QTL) assume that the phenotype follows a normal distribution and is fully observed. These assumptions may not be satisfied when the phenotype pertains to the survival time or failure time, which has a skewed distribution and is usually subject to censoring due to random loss of follow-up or limited duration of the experiment. In this article, we propose an interval-mapping approach for censored failure time phenotypes. We formulate the effects of QTL on the failure time through parametric proportional hazards models and develop efficient likelihood-based inference procedures. In addition, we show how to assess genome-wide statistical significance. The performance of the proposed methods is evaluated through extensive simulation studies. An application to a mouse cross is provided.

Chromosome Mapping↗