Inference about binocular sensitivity and specificity of screening tests for paired organs

Recently Perera et al. introduced two new binocular accuracy measures to evaluate diagnostic tests for paired organs. They adopted the Gaussian copula model to account for correlation between fellow eyes. As the measures are functions of several joint probabilities and due to the nature of the joint models, variations of the estimates for the two new measures were assessed via bootstrapping. We provide a different approach to inference about the two interesting and innovative measures. In our opinion, when patients are independent, the binomial models suffice for inference about the parameters of interest. Inference becomes simple and straightforward. We perform numerical studies and analyse the data set as of Perera et al. for illustration. Also, we investigate thru simulations the issue of robustness of the Gaussian copula and the binomial models under model misspecification.

Download Full-text

Robust likelihood inference for diagnostic accuracy measures for paired organs

Statistical Methods in Medical Research ◽

10.1177/0962280218798170 ◽

2018 ◽

Vol 28 (10-11) ◽

pp. 3163-3175 ◽

Cited By ~ 1

Author(s):

Tsung-Shan Tsou

Keyword(s):

Diagnostic Accuracy ◽

Likelihood Inference ◽

Paired Data ◽

Data Set ◽

Likelihood Approach ◽

The Common ◽

Accuracy Measures ◽

Robust Likelihood ◽

Fellow Eyes ◽

Joint Probabilities

Paired data arise naturally in Ophthalmology where pairs of eyes undergo diagnostic tests to predict the presence of certain diseases. The common correlation model is popular for modeling the joint probabilities of responses from fellow eyes for inference about accuracy measures. One of the assumptions underlying the model is exchangeability of fellow eyes that stipulates the accuracy measures such as sensitivities/specificities of fellow eyes be equal. We propose a parametric robust likelihood approach to testing the equality of accuracy measures of fellow eyes without modeling correlation. The robust likelihood procedure is applicable for inference about diagnostic accuracy measures in general paired designs. We provide simulations and analyses of a data set in Ophthalmology to demonstrate the effectiveness of the parametric robust procedure.

Download Full-text

Analysis of survey on menstrual disorder among teenagers using Gaussian copula model with graphical lasso prior

PLoS ONE ◽

10.1371/journal.pone.0248340 ◽

2021 ◽

Vol 16 (3) ◽

pp. e0248340

Author(s):

Jiali Wang ◽

Anton H. Westveld ◽

A. H. Welsh ◽

Melissa Parker ◽

Bronwyn Loong

Keyword(s):

Gaussian Copula ◽

Copula Model ◽

Data Set ◽

Menstrual Disorder ◽

Teenage Girls ◽

Graphical Lasso ◽

Cohort Differences ◽

Menstrual Disturbance ◽

Using Data ◽

Delays In Diagnosis

A high prevalence of menstrual disturbance has been reported among teenage girls, and research shows that there are delays in diagnosis of endometriosis among young girls. Using data from the Menstrual Disorder of Teenagers Survey (administered in 2005 and 2016), we propose a Gaussian copula model with graphical lasso prior to identify cohort differences in menstrual characteristics and to predict endometriosis. The model includes random effects to account for clustering by school, and we use the extended rank likelihood copula model to handle variables of mixed-type. The graphical lasso prior shrinks the elements in the precision matrix of a Gaussian distribution to encourage a sparse graphical structure, where the level of shrinkage is adaptable based on the strength of the conditional associations among questions in the survey. Applying our proposed model to the menstrual disorder data set, we found that menstrual disturbance was more pronouncedly reported over a decade, and we found some empirical differences between those girls with higher risk of developing endometriosis and the general population.

Download Full-text

Fast effect size shrinkage software for beta-binomial models of allelic imbalance

F1000Research ◽

10.12688/f1000research.20916.2 ◽

2020 ◽

Vol 8 ◽

pp. 2024

Author(s):

Joshua P. Zitovsky ◽

Michael I. Love

Keyword(s):

Allelic Imbalance ◽

Real Data ◽

Shrinkage Estimators ◽

Data Set ◽

Bayesian Shrinkage ◽

In Cis ◽

Posterior Estimation ◽

Binomial Models ◽

Better Than ◽

Diploid Organism

Allelic imbalance occurs when the two alleles of a gene are differentially expressed within a diploid organism and can indicate important differences in cis-regulation and epigenetic state across the two chromosomes. Because of this, the ability to accurately quantify the proportion at which each allele of a gene is expressed is of great interest to researchers. This becomes challenging in the presence of small read counts and/or sample sizes, which can cause estimators for allelic expression proportions to have high variance. Investigators have traditionally dealt with this problem by filtering out genes with small counts and samples. However, this may inadvertently remove important genes that have truly large allelic imbalances. Another option is to use pseudocounts or Bayesian estimators to reduce the variance. To this end, we evaluated the accuracy of four different estimators, the latter two of which are Bayesian shrinkage estimators: maximum likelihood, adding a pseudocount to each allele, approximate posterior estimation of GLM coefficients (apeglm) and adaptive shrinkage (ash). We also wrote C++ code to quickly calculate ML and apeglm estimates and integrated it into the apeglm package. The four methods were evaluated on two simulations and one real data set. Apeglm consistently performed better than ML according to a variety of criteria, and generally outperformed use of pseudocounts as well. Ash also performed better than ML in one of the simulations, but in the other performance was more mixed. Finally, when compared to five other packages that also fit beta-binomial models, the apeglm package was substantially faster and more numerically reliable, making our package useful for quick and reliable analyses of allelic imbalance. Apeglm is available as an R/Bioconductor package at http://bioconductor.org/packages/apeglm.

Download Full-text

A Data-Driven Exploratory Approach for Level Curve Estimation With Autonomous Underwater Agents

Volume 2: Mechatronics; Estimation and Identification; Uncertain Systems and Robustness; Path Planning and Motion Control; Tracking Control Systems; Multi-Agent and Networked Systems; Manufacturing; Intelligent Transportation and Vehicles; Sensors and Actuators; Diagnostics and Detection; Unmanned, Ground and Surface Robotics; Motion and Vibration Control Applications ◽

10.1115/dscc2017-5118 ◽

2017 ◽

Author(s):

Hsien-Chung Lin ◽

Eugen Solowjow ◽

Masayoshi Tomizuka ◽

Edwin Kreuzer

Keyword(s):

Concentration Field ◽

Real Data ◽

Support Vector ◽

Level Curve ◽

Data Set ◽

Curve Estimation ◽

Numerical Studies ◽

Exploratory Approach ◽

Vector Machines ◽

Myopic Strategy

This contribution presents a method to estimate environmental boundaries with mobile agents. The agents sample a concentration field of interest at their respective positions and infer a level curve of the unknown field. The presented method is based on support vector machines (SVMs), whereby the concentration level of interest serves as the decision boundary. The field itself does not have to be estimated in order to obtain the level curve which makes the method computationally very appealing. A myopic strategy is developed to pick locations that yield most informative concentration measurements. Cooperative operations of multiple agents are demonstrated by dividing the domain in Voronoi tessellations. Numerical studies demonstrate the feasibility of the method on a real data set of the California coastal area. The exploration strategy is benchmarked against random walk which it clearly outperforms.

Download Full-text

Rotation-Invariant texture retrieval using a steerable Gaussian copula model

2014 IEEE International Conference on Image Processing (ICIP) ◽

10.1109/icip.2014.7025603 ◽

2014 ◽

Author(s):

Hassan Rami ◽

Ahmed Drissi El Maliani ◽

Mohammed El Hassouni ◽

Yannick Berthoumieu

Keyword(s):

Gaussian Copula ◽

Copula Model ◽

Rotation Invariant ◽

Texture Retrieval

Download Full-text

Fast solution of the Gaussian copula model

Econometrics and Risk Management - Advances in Econometrics ◽

10.1016/s0731-9053(08)22001-3 ◽

2008 ◽

pp. 1-13 ◽

Cited By ~ 1

Author(s):

Bjorn Flesaker

Keyword(s):

Gaussian Copula ◽

Copula Model ◽

Fast Solution

Download Full-text

Moorfields AMD database report 2: fellow eye involvement with neovascular age-related macular degeneration

British Journal of Ophthalmology ◽

10.1136/bjophthalmol-2019-314446 ◽

2019 ◽

Vol 104 (5) ◽

pp. 684-690 ◽

Cited By ~ 3

Author(s):

Katrin Fasler ◽

Dun Jack Fu ◽

Gabriella Moraes ◽

Siegfried Wagner ◽

Eesha Gokhale ◽

...

Keyword(s):

Macular Degeneration ◽

Age Related Macular Degeneration ◽

Time Interval ◽

Data Set ◽

Fellow Eye ◽

Age Related ◽

Eye Involvement ◽

Baseline Characteristics ◽

Involvement Rate ◽

Fellow Eyes

Background/AimsNeovascular age-related macular degeneration (nAMD) is frequently bilateral, and previous reports on ‘fellow eyes’ have assumed sequential treatment after a period of treatment of the first eye only. The aim of our study was to analyse baseline characteristics and visual acuity (VA) outcomes of fellow eye involvement with nAMD, specifically differentiating between sequential and non-sequential (due to macular scarring in the first eye) antivascular endothelial growth factor treatment and timelines for fellow eye involvement.MethodsRetrospective, electronic medical record database study of the Moorfields AMD database of 6265 patients/120 286 single entries with data extracted between 21 October 2008 and 9 August 2018. The data set for analysis consisted of 1180 sequential, 807 non-sequential and 3410 unilateral eyes.ResultsMean VA (ETDRS letters±SD) of sequentially treated fellow eyes at baseline was significantly higher (63±13), VA gain over 2 years lower (0.37±14) and proportion of eyes with good VA (≥70 letters) higher (46%) than the respective first eyes (baseline VA 54±16, VA gain at 2 years 5.6±15, percentage of eyes with good VA 39%). Non-sequential fellow eyes showed baseline characteristics and VA outcomes similar to first eyes. Fellow eye involvement rate was 32% at 2 years, and median time interval to fellow eye involvement was 71 (IQR: 27–147) weeks.ConclusionThis report shows that sequentially treated nAMD fellow eyes have better baseline and final VA than non-sequentially treated eyes after 2 years of treatment. Sequentially treated eyes also had a greater proportion with good VA after 2 years.

Download Full-text

Empirical Models of Fish Response to Lake Acidification

Canadian Journal of Fisheries and Aquatic Sciences ◽

10.1139/f87-172 ◽

1987 ◽

Vol 44 (8) ◽

pp. 1432-1442 ◽

Cited By ~ 21

Author(s):

Kenneth H. Reckhow ◽

Robert W. Black ◽

Thomas B. Stockton Jr. ◽

J. David Vogt ◽

Judith G. Wood

Keyword(s):

New York ◽

Water Chemistry ◽

Brook Trout ◽

Yellow Perch ◽

Goodness Of Fit ◽

Lake Trout ◽

Model Misspecification ◽

Lake Acidification ◽

Data Set ◽

The Relationship

A large historical data set from the Adirondack region of New York was compiled to study the relationship between water chemistry variables associated with acid precipitation and the presence/absence of selected fish species. The data set was used to examine simple statistical models for fish presence/absence, as a function of the water chemistry variables, for brook trout (Salvelinus fontinalis), lake trout (Salvelinus namaycush), white sucker (Catostomus commersoni), and yellow perch (Perca flavescens). Of these models, only those for brook trout and lake trout were found to be acceptable based on statistical goodness-of-fit criteria; thus, parameters for models of these two species alone were estimated using maximum likelihood logistic regression. Candidate models for brook trout and lake trout were then examined, with particular consideration for the problems associated with model misspecification, errors-in-variables, and multicollinearity. For each of the two species, a model was recommended that may be used to predict the effect of changes in lake acidification on species presence/absence in lakes in the Adirondack region.

Download Full-text

Statistical evaluation of the influence of the uncertainty budget on B-spline curve approximation

Journal of Applied Geodesy ◽

10.1515/jag-2017-0018 ◽

2017 ◽

Vol 11 (4) ◽

Cited By ~ 4

Author(s):

Xin Zhao ◽

Hamza Alkhatib ◽

Boris Kargoll ◽

Ingo Neumann

Keyword(s):

Laser Scanning ◽

Statistical Significance ◽

Model Misspecification ◽

Uncertainty Budget ◽

Free Form ◽

Data Set ◽

Nested Model ◽

B Spline ◽

Uncertainty In Measurement ◽

Spline Curves

AbstractIn the field of engineering geodesy, terrestrial laser scanning (TLS) has become a popular method for detecting deformations. This paper analyzes the influence of the uncertainty budget on free-form curves modeled by B-splines. Usually, free-form estimation is based on scanning points assumed to have equal accuracies, which is not realistic. Previous findings demonstrate that the residuals still contain random and systematic uncertainties caused by instrumental, object-related and atmospheric influences. In order to guarantee the quality of derived estimates, it is essential to be aware of all uncertainties and their impact on the estimation.In this paper, a more detailed uncertainty budget is considered, in the context of the “Guide to the Expression of Uncertainty in Measurement” (GUM), which leads to a refined, heteroskedastic variance covariance matrix (VCM) of TLS measurements. Furthermore, the control points of B-spline curves approximating a measured bridge are estimated. Comparisons are made between the estimated B-spline curves using on the one hand a homoskedastic VCM and on the other hand the refined VCM. To assess the statistical significance of the differences displayed by the estimates for the two stochastic models, a nested model misspecification test and a non-nested model selection test are described and applied. The test decisions indicate that the homoskedastic VCM should be replaced by a heteroskedastic VCM in the direction of the suggested VCM. However, the tests also indicate that the considered VCM is still inadequate in light of the given data set and should therefore be improved.

Download Full-text