Bootstrap Confidence Intervals Research Articles

Classical Inference of a New PCI CNpmkc for Logistic-Exponential Process Distribution

The process capability index (PCI) has been developed as a useful instrument for measuring how well a product matches customer expectations while also assessing the performance of a process. We are aware that when the quality characteristics of the processes follow normal distribution, conventional PCIs produces superior results. However, these traditional indices could not yield reliable findings when assessing nonnormally distributed processes, which could make decision-making more difficult. In this paper, we take into account the [Formula: see text] process capacity index, which may be used for both normally distributed and nonnormally distributed processes. We have employed 10 traditional methods of estimation to estimate the PCI [Formula: see text] when the process has a logistic-exponential distribution, and the performances of these traditional estimates of the index [Formula: see text] are compared in terms of their mean squared errors through a simulation exercise. Then, we create five PCI [Formula: see text] bootstrap confidence intervals and contrast them based on their average widths and related coverage probabilities. In order to demonstrate the applicability of the suggested methods of estimation, two data sets pertaining to the electronic industries are re-analyzed.

Bootstrap Confidence Intervals for the Poisson-Pranav Distribution Parameter with an Application to COVID-19 Data

The Poisson distribution is commonly applied in statistical modeling to represent the count of events, assuming that the events are independent and occur at a constant rate. This assumption, however, may not always hold true in real-life situations. The Poisson distribution may be inadequate if the underlying rate of occurrence is not constant. The mixed Poisson distribution is proposed to solve this limitation because it permits the rate parameter of the Poisson distribution to be random rather than fixed. The Poisson-Pranav distribution, which is classified as a type of mixed Poisson distribution, has been commonly utilized to analyze count data that exhibits over-dispersion over many domains. However, no research has been done into building confidence intervals for the parameter of the Poisson-Pranav distribution using the bootstrap method. Using Monte Carlo simulation, the coverage probabilities and average lengths of the percentile bootstrap (PB), basic bootstrap (BB), and biased-corrected and accelerated (BCa) bootstrap’s interval-estimation performances were compared. The bootstrap method was not appropriate for achieving the desired nominal confidence level with small sample sizes. In addition, the performance of the bootstrap confidence intervals did not differ significantly when the sample size was increased considerably. In each case studied, the BCa bootstrap confidence intervals performed better than the others. The effectiveness of bootstrap confidence intervals was demonstrated by applying them to the number of COVID-19 deaths in Belgium. The computations substantially supported the proposed bootstrap confidence intervals.

Smoothed gradient least squares estimator for linear threshold models

In the presence of fixed threshold effects, the least squares (LS) estimator of the threshold parameter poses challenges for statistical inference due to its nonstandard limiting distribution, which also presents challenges for bootstrap methods. To address this issue, we propose a novel estimator: a two-step smoothed gradient least squares (SGLS) estimator. Our proposed method achieves a normal limiting distribution for the threshold parameter with minimal efficiency loss compared to the LS estimator. Furthermore, our modified bootstrap method significantly enhances computational efficiency, leading to improved bootstrap confidence intervals (CIs) for the threshold parameter compared to asymptotic CIs. Our method is validated through a small Monte Carlo study and demonstrated with an empirical application.

Abstract PO5-14-08: External Reproducibility of PD-L1 IHC 22C3 pharmDx on Breast Cancer Specimens at CPS ≥ 1 and CPS ≥ 10 Cutoffs

Abstract Background: PD-L1 IHC 22C3 pharmDx is an approved companion diagnostic for pembrolizumab treatment in triple-negative breast cancer (TNBC). Clinical evidence for pembrolizumab approval in TNBC was demonstrated by KEYNOTE-355 (NCT02819518), which investigated the clinical validity of PD-L1 IHC 22C3 pharmDx in identifying patients with PD-L1 expressing Combined Positive Score (CPS) ≥ 10 in previously untreated locally recurrent unresectable or metastatic TNBC.1 The role of immune checkpoint inhibitors for breast cancer as a whole is not as well defined as it is for TNBC. In this study, we show evidence of the reproducibility of PD-L1 IHC 22C3 pharmDx in PD-L1 expression determination in breast cancer specimens at the CPS ≥ 1 and CPS ≥ 10 cutoffs documenting robust assay performance in breast cancer. Methods: A blinded, randomized study was conducted at three external sites using formalin-fixed, paraffin-embedded breast cancer specimens. Reproducibility of PD-L1 IHC 22C3 pharmDx was assessed at the inter-site, intra-site/inter-day, inter-observer, and intra-observer endpoints at the CPS ≥ 1 and CPS ≥ 10 cutoffs. To assess inter-site and intra-site reproducibility, five replicate sets of pre-qualified unstained specimens were stained and evaluated at each of the three external sites. To assess inter-observer and intra-observer reproducibility, one set of pre-stained specimens was evaluated three times by each of three external pathologists. Negative percent agreement (NPA), positive percent agreement (PPA), and overall percent agreement (OA), based on comparisons to consensus diagnostic outcome, were computed with corresponding two-sided 95% percentile bootstrap confidence intervals. Acceptance criteria required at least 85.0% for the lower-bound (LB) of the two-sided 95% confidence interval (CI) on NPA, PPA, and OA for all endpoints. Results: All but one endpoint in the external reproducibility study met pre-determined acceptance criteria. Both the inter- and intra-site reproducibility endpoints had 95% CI LB values of 98.8% for NPA, 95.9% for PPA, and 97.9% OA at the CPS ≥ 1 cutoff. At the CPS ≥ 10 cutoff, the inter-site reproducibility endpoint had 95% CI LB values of 88.0% for NPA, 87.1% for PPA, and 89.5% OA, and the intra-site reproducibility endpoint had 95% CI LB values of 96.7% for NPA, 92.2% for PPA, and 95.6% OA. The inter-observer reproducibility endpoint had 95% CI LB values of 95.3% for NPA, 92.3% for PPA, and 94.7% OA at the CPS ≥ 1 cutoff. At the CPS ≥ 10 cutoff, the inter-observer reproducibility endpoint had 95% CI LB values of 84.9% for NPA, 86.8% for PPA, and 87.3% OA. The intra-observer reproducibility endpoint had 95% CI LB values of 97.0% for NPA, 95.0% for PPA, and 96.6% OA at the CPS ≥ 1 cutoff. At the CPS ≥ 10 cutoff, the intra-observer reproducibility endpoint had 95% CI LB values of 92.2% for NPA, 89.2% for PPA, and 91.7% OA. Conclusions: Acceptance criteria were met for all endpoints in the breast cancer external reproducibility study of PD-L1 IHC 22C3 pharmDx at the CPS ≥ 1 and CPS ≥ 10 cutoffs, except for inter-observer at the CPS ≥ 10 cutoff, which had a 95% CI LB value of 84.9% for NPA.

Proposal for a classification system of radiographic bone changes after cervical disc replacement

BackgroundThe goal of this study is to propose a classification system with a common nomenclature for radiographic observations of periprosthetic bone changes following cTDR.MethodsAided by serial plain radiographs from recent cTDR cases (34 patients; 44 devices), a panel of experts assembled for the purpose of creating a classification system to aid in reproducibly and accurately identifying bony changes and assessing cTDR radiographic appearance. Subdividing the superior and inferior vertebral bodies into 3 equal sections, observed bone loss such as endplate rounding, cystic erosion adjacent to the endplate, and cystic erosion not adjacent to the endplate, is recorded. Determining if bone loss is progressive, based on serial radiographs, and estimating severity of bone loss (measured by the percentage of end plate involved) is recorded. Additional relevant bony changes and device observations include radiolucent lines, heterotopic ossification, vertebral body olisthesis, loss of core implant height, and presence of device migration, and subsidence.ResultsSerial radiographs from 19 patients (25 devices) implanted with a variety of cTDR designs were assessed by 6 investigators including clinicians and scientists experienced in cTDR or appendicular skeleton joint replacement. The overall agreement of assessments ranged from 49.9% (95% bootstrap confidence interval 45.1–73.1%) to 94.7% (95% CI 86.9–100.0%). There was reasonable agreement on the presence or absence of bone loss or radiolucencies (range: 58.4% (95% CI 51.5–82.7%) to 94.7% (95% CI 86.9–100.0%), as well as in the progression of radiolucent lines (82.9% (95% CI 74.4–96.5%)).ConclusionsThe novel classification system proposed demonstrated good concordance among experienced investigators in this field and represents a useful advancement for improving reporting in cTDR studies.

On Sample Size Needed for Block Bootstrap Confidence Intervals to Have Desired Coverage Rates

Block bootstrap is widely used in constructing confidence intervals for parameters estimated from stationary time series. Theoretically, the method should provide valid confidence intervals as the length of the time series goes to infinity. In practice, however, it is necessary to know how large of a finite sample is required for block bootstrap confidence intervals to work well. This study aims to answer this question in a simple simulation setting where the data are generated from a first-order autoregressive process. The empirical coverage rates of several commonly used bootstrap confidence intervals for the mean, standard deviation, and the lag-1 autocorrelation coefficient are compared. A quite large sample is found necessary for the intervals to have the right coverage rates even when estimating a simple parameter like the mean. Some block bootstrap methods could fail when estimating the lag-1 autocorrelation. It is surprising that the coverage property even deteriorates as the sample size increases with some commonly used block bootstrap confidence intervals including the percentile intervals and bias-corrected intervals. KEYWORDS: Autocorrelation; Bias-Correction; Centering; Dependent Data; Percentile; Resampling; Simulation; Time Series

Constant-Stress Partially Accelerated Life Testing for Weibull Inverted Exponential Distribution with Censored Data

The novelty of this article is estimating the parameters of Weibull inverted exponential (WIE)distribution with a constant stress partially accelerated life test (PALT) under adaptive type-II progressivelycensored samples. Moreover, the maximum likelihood estimators (MLEs), their asymptotic variances confidenceintervals, and Bayes estimators (BEs) of the model parameters and the acceleration factor are obtained.Furthermore, the approximate bootstrap and credible confidence intervals of the estimators are acquired. Theaccuracy of the MLEs and BEs for the model parameters and the acceleration factor is investigated through thesimulation studies.

Biomarker and proteome analysis of milk from dairy cows with clinical mastitis: Determining the effect of different bacterial pathogens on the response to infection

Antimicrobial usage (AMU) could be reduced by differentiating the causative bacteria in cases of clinical mastitis (CM) as either Gram-positive or Gram-negative bacteria or identifying whether the case is culture-negative (no growth, NG) mastitis. Immunoassays for biomarker analysis and a Tandem Mass Tag (TMT) proteomic investigation were employed to identify differences between samples of milk from cows with CM caused by different bacteria. A total of 94 milk samples were collected from cows diagnosed with CM across seven farms in Scotland, categorized by severity as mild (score 1), moderate (score 2), or severe (score 3). Bovine haptoglobin (Hp), milk amyloid A (MAA), C-reactive protein (CRP), lactoferrin (LF), α-lactalbumin (LA) and cathelicidin (CATHL) were significantly higher in milk from cows with CM, regardless of culture results, than in milk from healthy cows (all P-values <0.001). Milk cathelicidin (CATHL) was evaluated using a novel ELISA technique that utilises an antibody to a peptide sequence of SSEANLYRLLELD (aa49–61) common to CATHL 1–7 isoforms. A classification tree was fitted on the six biomarkers to predict Gram-positive bacteria within mastitis severity scores 1 or 2, revealing that compared to the rest of the samples, Gram-positive samples were associated with CRP < 9.5 μg/ml and LF ≥ 325 μg/ml and MAA < 16 μg/ml. Sensitivity of the tree model was 64%, the specificity was 91%, and the overall misclassification rate was 18%. The area under the ROC curve for this tree model was 0.836 (95% bootstrap confidence interval: 0.742; 0.917). TMT proteomic analysis revealed little difference between the groups in protein abundance when the three groups (Gram-positive, Gram-negative and no growth) were compared, however when each group was compared against the entirety of the remaining samples, 28 differentially abundant protein were identified including β-lactoglobulin and ribonuclease. Whilst further research is required to draw together and refine a suitable biomarker panel and diagnostic algorithm for differentiating Gram- positive/negative and NG CM, these results have highlighted a potential panel and diagnostic decision tree. Host-derived milk biomarkers offer significant potential to refine and reduce AMU and circumvent the many challenges associated with microbiological culture, both within the lab and on the farm, while providing the added benefit of reducing turnaround time from 14 to 16 h of microbiological culture to just 15 min with a lateral flow device (LFD).

Abstract 3754: High staining reproducibility and scoring precision demonstrated in endometrial cancer stained with PD-L1 IHC 22C3 pharmDx used in conjunction with combined positive score

Abstract Background: PD-L1 IHC 22C3 pharmDx (SK006) is a qualitative immunohistochemical (IHC) assay using anti-PD-L1 clone 22C3 to detect PD-L1 in formalin-fixed, paraffin-embedded (FFPE) human endometrial cancer (ENDO) tissues. PD-L1 expression is determined using the Combined Positive Score (CPS) algorithm, which is the number of PD-L1 staining cells (tumor cells, lymphocytes, macrophages) divided by the total number of viable tumor cells, multiplied by 100. Validation testing evaluated the sensitivity and analytical performance of SK006 PD-L1 expression in FFPE ENDO specimens based on two PD-L1 expression cutoffs for all applicable studies: CPS ≥ 1 and CPS ≥ 10. Analytical performance studies included precision (combined precision, intra-run and intra-inter observer) and robustness (time, temperature, pH and re-use) testing. Methods: FFPE ENDO specimens were stained with SK006 and scored using CPS. Negative percent agreement (NPA), positive percent agreement (PPA), and overall agreement (OA) with two-sided 95% bootstrap confidence intervals (CIs) were used for data analysis, based on PD-L1 binary status (positive/negative) at the evaluated cutoffs. Acceptance Criteria (AC): the lower bound of the two-sided 95% CI computed on percent agreement (NPA, PPA, and OA) must meet or exceed 85%. Sensitivity determined if SK006 could detect a range of CPS expression levels in ENDO specimens. Combined precision measured inter-instrument, -operator, and -day variation nested within the same study. Intra-run measured repeatability within the same run and Inter/Intra-Observer measured scoring reproducibility by 3 pathologists over 3 non-consecutive days. Robustness evaluated the staining performance of SK006 under varying Target Retrieval Solution (TRS) laboratory conditions: time (18-22min), temperature (95-99°C), pH (5.9-6.3) and re-use (up to 3 uses). Results: The sensitivity data showed that SK006 could detect a range of CPS expression levels in ENDO specimens. Analyses for each analytical validation study, subject to AC, showed NPA, PPA, OA point estimates (PE) of ≥ 95.5% and CI lower bounds (CILB) of ≥ 87.1%. Conclusion: PD-L1 IHC 22C3 pharmDx (SK006) used in conjunction with CPS provides high staining reproducibility and scoring precision for ENDO specimens at the CPS ≥ 1 and CPS ≥ 10 cutoffs, under standard day-to-day laboratory testing and scoring conditions. Citation Format: Julia Hand, Amber Morones, Deanna Moquin, Kelly Martyniuk, Megan Kalpakoff, Siena Tabuena-Frolli, Jay Milo, Stephanie Hund. High staining reproducibility and scoring precision demonstrated in endometrial cancer stained with PD-L1 IHC 22C3 pharmDx used in conjunction with combined positive score [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3754.

The relationship between expressive suppression, therapeutic bond, and treatment outcomes of a positive affect intervention for adults with anxiety and/or depression

ABSTRACT Expressive suppression (ES; reducing emotional expression) is linked with reduced social connectedness in individuals with anxiety or depression. One implication is that people who use ES may have difficulty establishing a bond with their therapist which may impede clinical improvement. We examined this hypothesis in 33 adults with clinically elevated anxiety or depression receiving treatment focused on enhancing positive thoughts, emotions, and behaviors. At baseline, participants rated ES for positive and negative emotions during a standardized conversation task designed to generate connectedness. They also rated measures of early (session 3) perceived therapeutic bond and treatment outcomes (i.e. positive affect and social connectedness). ES of positive (r = -.39, p = .018), but not negative (r = .06, p = .747), emotions was negatively associated with therapeutic bond. Therapeutic bond mediated the relationship between greater ES of positive emotions during affiliation and lower post-treatment positive affect, 95% bias-corrected bootstrap confidence interval [−0.021, −0.000], adjusted for pre-treatment positive affect, as well as lower post-treatment social connectedness [−0.397, −0.015]; however, the indirect effect was not significant when accounting for pre-treatment social connectedness (p > .05). ES of positive emotions may be an important factor in the development of therapeutic bond and therefore treatment outcomes for individuals with anxiety or depression.

The effects of clinical learning environment and career adaptability on resilience: A mediating analysis based on a survey of nursing interns.

The resilience education of intern nursing students has significant implications for the development and improvement of the nursing workforce. The clinical internship period is a critical time for enhancing resilience. To evaluate the resilience level of Chinese nursing interns and explore the effects of factors affecting resilience early in their careers, focusing on the mediating roles of career adaptability between clinical learning environment and resilience. The cross-sectional study design was adopted. From March 2022 to May 2023, 512 nursing interns in tertiary care hospitals were surveyed online with the Connor-Davidson Resilience Scale, the Clinical Learning Environment Scale for Nurse and the Career Adapt-Abilities Scale. Structural equation modelling was used to clarify the relationships among these factors. Indirect effects were tested using bootstrapped confidence intervals. The nursing interns showed a moderately high level of resilience [M (SD) = 70.15 (19.90)]. Gender, scholastic attainment, scholarship, career adaptability and clinical learning environment were influencing factors of nursing interns' resilience. Male interns with good academic performance showed higher levels of resilience. Career adaptability and clinical learning environment positively and directly affected their resilience level (β = 0.62, 0.18, respectively, p < .01). Career adaptability was also positively affected by the clinical learning environment (β = 0.36, p < .01), and mediated the effect of clinical learning environment on resilience (β = 0.22, p < .01). Clinical learning environment can positively affect the resilience level of nursing interns. Career adaptability can affect resilience directly and also play a mediating role between clinical learning environment and resilience. Thus, promotion of career adaptability and clinical teaching environment should be the potential strategies for nursing interns to improve their resilience, especially for female nursing interns with low academic performance.

Confidence intervals for validation statistics with data truncation in genomic prediction

BackgroundValidation by data truncation is a common practice in genetic evaluations because of the interest in predicting the genetic merit of a set of young selection candidates. Two of the most used validation methods in genetic evaluations use a single data partition: predictivity or predictive ability (correlation between pre-adjusted phenotypes and estimated breeding values (EBV) divided by the square root of the heritability) and the linear regression (LR) method (comparison of “early” and “late” EBV). Both methods compare predictions with the whole dataset and a partial dataset that is obtained by removing the information related to a set of validation individuals. EBV obtained with the partial dataset are compared against adjusted phenotypes for the predictivity or EBV obtained with the whole dataset in the LR method. Confidence intervals for predictivity and the LR method can be obtained by replicating the validation for different samples (or folds), or bootstrapping. Analytical confidence intervals would be beneficial to avoid running several validations and to test the quality of the bootstrap intervals. However, analytical confidence intervals are unavailable for predictivity and the LR method.ResultsWe derived standard errors and Wald confidence intervals for the predictivity and statistics included in the LR method (bias, dispersion, ratio of accuracies, and reliability). The confidence intervals for the bias, dispersion, and reliability depend on the relationships and prediction error variances and covariances across the individuals in the validation set. We developed approximations for large datasets that only need the reliabilities of the individuals in the validation set. The confidence intervals for the ratio of accuracies and predictivity were obtained through the Fisher transformation. We show the adequacy of both the analytical and approximated analytical confidence intervals and compare them versus bootstrap confidence intervals using two simulated examples. The analytical confidence intervals were closer to the simulated ones for both examples. Bootstrap confidence intervals tend to be narrower than the simulated ones. The approximated analytical confidence intervals were similar to those obtained by bootstrapping.ConclusionsEstimating the sampling variation of predictivity and the statistics in the LR method without replication or bootstrap is possible for any dataset with the formulas presented in this study.

Alexithymia and negative emotions among nursing students: a moderated mediation model

Nursing students, who comprise a high percentage of China’s college students, experience many psychological problems; however, few studies explored the mechanisms underlying these problems. This cross-sectional study explored the relationships and mechanisms of depression, anxiety, stress, and narrative disorders in senior nursing students. Questionnaires were administered to 380 senior nursing students in Hubei Province using the Sociodemographic Questionnaire, Toronto Alexithymia-20 Scale, Perceived Social Support Scale, 10-Item Connor-Davidson Resilience Scale, and Depression-Anxiety-Stress Scale. After controlling for sociodemographic variables, Hayes’ PROCESS macros were used to test how psychological resilience moderates the relationships among narrative disorders, negative affect, and perceived social support. Bootstrap confidence intervals tested for indirect effects. Correlation analyses revealed that alexithymia was correlated significantly positively with depression-anxiety-stress (r = 0.57, 0.56, and 0.58, resp.) and significantly negatively with perceived social support (r = 0-0.46). Psychological resilience was correlated significantly negatively with alexithymia (r=-0.39) and depression-anxiety-stress (r=-0.31, -0.30, and-0.32, resp.) but significantly positively with perceived social support(r = 0.50). Perceived social support was correlated significantly negatively with depression-anxiety-stress (r=-0.33, -0.34, and − 0.42 resp.). Stress was correlated significantly positively with anxiety and depression (r = 0.81 and 0.77, resp.). Psychological resilience was a partial mediator between depression and dysphoria (β=-0.08, p < 0.05). Dysphoria directly predicted anxiety (β = 0.31) and stress (β = 0.37); moreover,alexithymia predicted depression not only directly but also through the mediating effect of psychological resilience. Therefore, educators and clinical administrators must promote and recognise negative emotions among nursing students to help ensure the nursing workforce’s stability.

A generating family of unit-Garima distribution: Properties, likelihood inference, and application

This article proposes the unit Garima (UGa) distribution for analysing proportion data. Some statistical properties of the UGa distribution are investigated, including survival and hazard functions, order statistics, quantile function, and stress-strength reliability measure. Next, a new family of continuous distributions, called the unit Garima-generated (UGa-G) family of distributions, is studied. The UGa-G family of distributions has the feature to use the UGa distribution as the main generator and the concept of the T-X family of distributions. Some UGa-G family sub-models are provided, such as the UGa-Beta, UGa-Weibull, and UGa-normal distributions. The maximum likelihood method is used to estimate the model parameters for the statistical aspect. A Monte Carlo simulation for the percentile bootstrap confidence intervals for each parameter of the proposed distributions is provided. Applications to eight practical data sets are given to demonstrate the usefulness of the proposed distributions.

Estimation of quantiles with the exact bootstrap method

A problem with the estimation of quantiles occurs when the sample comes from an unknown distribution. The estimation uses the bootstrap method in the version that the literature refers to as exact. Three bootstrap estimators were used: two of them based on one order statistic, and the third on a linear combination of two order statistics (for an integer). The distribution of the exact bootstrap estimator based on a single order statistic is known. It has been shown that there is no general form of the distribution of the exact bootstrap estimator based on two order statistics. However, it is possible to calculate such a distribution – the article presents the algorithm that performs such a task. The bootstrap confidence intervals were constructed using the exact percentile method. It has been shown that if the estimator is based on a single order statistic, it is known in advance which elements of the primary sample are the limits of the confidence intervals, so there is no need to resample. The intervals determined by the exact percentile method were compared with those constructed using other methods. It has been shown that the information on the direction of the asymmetry of the distribution that the sample comes from is worth considering when selecting the rank of the order statistic used as an estimator. Attention is paid to the influence of the quality of the pseudorandom number generators on the results of the Monte Carlo simulation.

BOOTSTRAP CONFIDENCE INTERVALS IN LINEAR MODELS: CASE OF OUTLIERS

Confidence interval estimations in linear models have been of large interest in social science. However, traditional approach of building confidence intervals has a set of assumption including dataset having no extreme outliers. In this study, we discuss presence of severe outliers in linear models and suggest bootstrap approach as an alternative way to construct confidence intervals. We conclude that bootstrap confidence intervals can outperform traditional confidence intervals in presence of outliers when sample size is small or population distribution is not normal. Lastly, we encourage researchers to run a computer simulation to evaluate conclusions of this study.

On the study of the stress-strength reliability in Weibull-$F$ Models

In this paper, the problem of inferencing on the stress-strength reliability under Weibull-$F$ Models when the stress and strength systems belong to the different families of distributions from the Weibull-$F$ Model is considered. Some stochastic comparisons between the survival distribution functions of this model are obtained. Also, the asymptotic and several bootstrap confidence intervals of stress-strength reliability are studied. The efficiency of asymptotic and bootstrap confidence intervals are analyzed by simulation. The numerical example based on real-life data is displayed as an illustration.

The moderating role of perceived brand globalness on the effects of consumers’ attitude during periods of political controversies: three interconnected studies following marketing analytics’ best practices

AbstractThe purpose of this research is to test the moderating role of perceived brand globalness (PBG) on consumer unfavourable attitudes and buying intention towards brands associated with a conflicting political belief. Three studies were conducted. Study 1 (N = 228) examines the effect of a brand being associated with a conflicting political belief on attitude and buying intention. Study 2 (N = 212) tests the moderating role of PBG among these relationships. Study 3 (N = 126) tests the moderating role of PBG for a service brand and adds the mediating role of self-congruity. We followed marketing analytics best practices to examine causal effects of manipulated stimuli. We also examined the psychometric characteristics of the data-collection tool and tested direct relationships between variables. Additionally, we used percentile bootstrap confidence intervals, with PROCESS-Macro, to test mediating effects in the theoretical model. The studies include moderating analyses and analysis of variance. The results show that the association of a brand with a conflicting political belief negatively influences perceived self-congruity, which leads to an unfavourable attitude and, in turn, to a lower buying intention. Presenting a brand as global mitigates such unfavourable effects. The moderating effect of PBG is confirmed for a tangible product and for a service brand.

Omics feature selection with the extended SIS R package: identification of a body mass index epigenetic multi-marker in the Strong Heart Study.

The statistical analysis of omics data poses a great computational challenge given its ultra-high dimensional nature and frequent between-features correlation. In this work, we extended the Iterative Sure Independence Screening (ISIS) algorithm by pairing ISIS with elastic-net (Enet) and two versions of adaptive Enet (AEnet and MSAEnet) to efficiently improve feature selection and effect estimation in omics research. We subsequently used genome-wide human blood DNA methylation data from American Indians of the Strong Heart Study (N=2,235 participants), measured in 1989-1991, to compare the performance (predictive accuracy, coefficient estimation and computational efficiency) of SIS-paired regularization methods to Bayesian shrinkage and traditional linear regression to identify epigenomic multi-marker of body mass index. ISIS-AEnet outperformed the other methods in prediction. In biological pathway enrichment analysis of genes annotated to BMI-related differentially methylated positions, ISIS-AEnet captured most of the enriched pathways in common for at least two of all the evaluated methods. ISIS-AEnet can favor biological discovery because it identifies the most robust biological pathways while achieving an optimal balance between bias and efficient feature selection. In the extended SIS R package, we also implemented ISIS paired with Cox and logistic regression for time-to-event and binary endpoints, respectively, and bootstrap confidence intervals for the estimated regression coefficients.

Linear mixed models and latent growth curve models for group comparison studies contaminated by outliers.

The linear mixed model (LMM) and latent growth model (LGM) are frequently applied to within-subject two-group comparison studies to investigate group differences in the time effect, supposedly due to differential group treatments. Yet, research about LMM and LGM in the presence of outliers (defined as observations with a very low probability of occurrence if assumed from a given distribution) is scarce. Moreover, when such research exists, it focuses on estimation properties (bias and efficiency), neglecting inferential characteristics (e.g., power and type-I error). We study power and type-I error rates of Wald-type and bootstrap confidence intervals (CIs), as well as coverage and length of CIs and mean absolute error (MAE) of estimates, associated with classical and robust estimations of LMM and LGM, applied to a within-subject two-group comparison design. We conduct a Monte Carlo simulation experiment to compare CIs and MAEs under different conditions: data (a) without contamination, (b) contaminated by within-subject outliers, (c) contaminated by between-subject outliers, and (d) both contaminated by within- and between-subject outliers. Results show that without contamination, methods perform similarly, except CIs based on S, a robust LMM estimator, which are slightly less close to nominal values in their coverage. However, in the presence of both within- and between-subject outliers, CIs based on robust estimators, especially S, performed better than those of classical methods. In particular, the percentile CI with the wild bootstrap applied to the robust LMM estimators outperformed all other methods, especially with between-subject outliers, when we found the classical Wald-type CI based on the t statistic with Satterthwaite approximation for LMM to be highly misleading. We provide R code to compute all methods presented here. (PsycInfo Database Record (c) 2024 APA, all rights reserved).