Proper Confidence Research Articles

Improving the Estimation of Prediction Increment Measures in Logistic and Survival Analysis.

Background/Objectives: Proper confidence interval estimation of the area under the receiver operating characteristic curve (AUC), the net reclassification index (NRI), and the integrated discrimination improvement (IDI) is an area of ongoing research. The most common confidence interval estimation methods employ asymptotic theory. However, developments demonstrate that degeneration of the normal distribution assumption under the null hypothesis exists for measures such as the change in AUC (ΔAUC) and IDI, and confidence intervals estimated under the normal distribution assumption may be invalid. We aim to study the performance of confidence intervals derived assuming asymptotic theory and those derived with non-parametric bootstrapping methods. Methods: We examine the performance of ΔAUC, NRI, and IDI in both the logistic and survival regression context. We explore empirical distributions and compare coverage probabilities of asymptotic confidence intervals with those produced from bootstrapping methods through simulation. Results: The primary finding in both the logistic framework and the survival analysis framework is that the percentile CIs performed well regarding coverage, without compromise to their width; this finding was robust in most scenarios. Conclusions: Our results suggest that the asymptotic intervals are only appropriate when a strong effect size of the added parameter exists, and that the percentile bootstrap interval exhibits at least a reasonable coverage while maintaining the shortest width in nearly all simulated scenarios, making this interval the most reliable choice. The intent is that these recommendations improve the accuracy in the estimation and the overall assessment of discrimination improvement.

The similative-pretence alternating pair and filler-slot relations

Abstract The present paper investigates the similative-pretence alternating pair in Mexican Spanish (como/igual que and como si constructions) based on the analysis of 1362 instances from The Corpus del Español NOW (News on the Web) corpus. We apply a revised version of distinctive collexeme analysis to our concordance data to explore the variety of different verbs that can occur in the first slot of similative and pretence constructions. Specifically, we use (i) a new measure that distinguishes the attraction that a verb lemma exerts on a construction from the attraction that a construction exerts on a verb lemma and (ii) bootstrapping on the level of files to, for the first time, provide proper confidence intervals in the context of collostructional studies. The from-verb lemma-to-construction analysis shows that there is a significant attraction of epistemic lemmas (e.g., parecer ‘to seem’) to occur in como ‘like’, and of mistaken identity lemmas (e.g., actuar ‘to act’) to appear in como si ‘as if’ constructions. The from-construction-to-verb lemma analysis, on the other hand, demonstrates that igual que ‘like’ significantly attracts epistemic perception verbs (e.g., ver ‘to look’), and como si ‘as if’ constructions significantly attract the verb lemma sentir ‘to feel.’

On Certainty, Proper Confidence, and Doubt: A Postfoundationalist and Critical Realist Epistemology for Doing Constructive Theology

ABSTRACTThe present essay1 seeks to focus on the meaning and challenges of doubt and lack of certainty in the context of doing academic theological work, whether in a Seminary setting such as my own, Fuller Theological Seminary, or in a “secular” university setting such as the University of Helsinki, my other institution. Throughout, I am raising questions such as whether any kind of certainty or confidence is available in the work of theology, particularly constructive theology, my own particular area of interest. Speaking of constructive theology, reflecting on the issues of doubt, certainty, and confidence is particularly important for the simple reason that it typically engages some non‐theological disciplines such as natural sciences with their own criteria for truth. And in my case, dialogue with some other faith traditions, including Islam, further makes the issue of doubt pertinent. To make my inquiry manageable—and hopefully also dynamic—I am carrying critical conversation with two thinkers of the 20th century, namely the philosopher Ludwig Wittgenstein and the missionary bishop and cultural critic of the West, Lesslie Newbigin. Whereas the mature Wittgenstein was in search of certainty towards the end of his life, Newbigin sought for proper confidence. Both of them rejected the Cartesian indubitable certainty of modernity, on the one hand, and the dismissal of the possibility of truth in many streams of late modernity on the other. While critically affirming some key arguments of these two diverse thinkers, I also see the importance of buttressing their emerging epistemological programs with the help of insights from postfoundationalist and critical–realist epistemologies. It is clear without saying that this big task can only be outlined and sketched rather than scrutinized in any detail within a 1‐h talk! My presentation is more an invitation to conversation and less any kind of finished project. I will first engage the two protagonists. Thereafter, before the summative and concluding reflections in terms of implications to theology and faith, a brief look at the meaning and nature of postfoundationalism and critical realism is attempted.

Exact Policy Recovery in Offline RL with Both Heavy-Tailed Rewards and Data Corruption

We study offline reinforcement learning (RL) with heavy-tailed reward distribution and data corruption: (i) Moving beyond subGaussian reward distribution, we allow the rewards to have infinite variances; (ii) We allow corruptions where an attacker can arbitrarily modify a small fraction of the rewards and transitions in the dataset. We first derive a sufficient optimality condition for generalized Pessimistic Value Iteration (PEVI), which allows various estimators with proper confidence bounds and can be applied to multiple learning settings. In order to handle the data corruption and heavy-tailed reward setting, we prove that the trimmed-mean estimation achieves the minimax optimal error rate for robust mean estimation under heavy-tailed distributions. In the PEVI algorithm, we plug in the trimmed mean estimation and the confidence bound to solve the robust offline RL problem. Standard analysis reveals that data corruption induces a bias term in the suboptimality gap, which gives the false impression that any data corruption prevents optimal policy learning. By using the optimality condition for the generalized PEVI, we show that as long as the bias term is less than the ``action gap'', the policy returned by PEVI achieves the optimal value given sufficient data.

A Proper Confidence: The Prophetic Nature of Adolescent Discipleship

Abstract This paper argues that to help adolescents respond to Jesus’ call to pick up their cross and follow him, the contemporary Pentecostal/Charismatic practice of prophecy/revelatory experience must be reimagined and framed within the ministry of Christ. Without this Christocentric framework these experiences are reduced to immediate therapeutic experiences of individual flourishing and are not sufficient to redefine the adolescent’s past and offer direction for a future developing faith identity. Synthesis of Walter Brueggemann’s The Prophetic Imagination, with adolescent identity formation and critique of the contemporary Pentecostal practice of prophecy highlights implications for realigned practice in order to foster the confidence in adolescent believers to form counter cultural identities in light of the Coming Kingdom.

Confidence set-based analysis of minimal detectable fault under hybrid Gaussian and bounded uncertainties

Confidence set-based analysis of minimal detectable fault under hybrid Gaussian and bounded uncertainties

Inference on subgroups identified based on a heterogeneous treatment effect in a post hoc analysis of a clinical trial.

Due to the many benefits of understanding treatment effect heterogeneity in a clinical trial, an exploratory post hoc subgroup analysis is often performed to find subpopulations of patients with conditional average treatment effect that suggests better treatment efficacy than in the overall population. A naive re-substitution approach uses all available data to identify a subgroup and then proceeds with estimation and inference using the same data set. This approach generally leads to an overly optimistic estimate of conditional average treatment effect. In this article, in a post hoc analysis, we estimate the target optimal subgroup through maximizing a utility function, from candidates systematically identified with a penalized regression. We then compare two resampling-based bias-correction methods, cross-validation and debiasing bootstrap, for obtaining approximately unbiased estimates and valid inference of conditional average treatment effect in the identified subgroup, with either an empirical or an augmented estimator. Our results show that both the cross-validation and the debiasing bootstrap methods reduce the re-substitution bias effectively. The cross-validation method appears to have less biased point estimates, smaller standard error estimates, but poorer coverages than the debiasing bootstrap method when using the empirical estimator and the sample size is moderate. Using the augmented estimator in the debiasing bootstrap method leads to less biased point estimates but poorer coverages. We conclude that bias correction should be a part of every exploratory post hoc subgroup analysis to eliminate re-substitution bias and to obtain a proper confidence interval for the estimated conditional average treatment effect in the selected subgroup.

Machine Learning for Exposure-Response Analysis: Methodological Considerations and Confirmation of Their Importance via Computational Experimentations.

Exposure-response (E-R) is a key aspect of pharmacometrics analysis that supports drug dose selection. Currently, there is a lack of understanding of the technical considerations necessary for drawing unbiased estimates from data. Due to recent advances in machine learning (ML) explainability methods, ML has garnered significant interest for causal inference. To this end, we used simulated datasets with known E-R "ground truth" to generate a set of good practices for the development of ML models required to avoid introducing biases when performing causal inference. These practices include the use of causal diagrams to enable the careful consideration of model variables by which to obtain desired E-R relationship insights, keeping a strict separation of data for model-training and for inference generation to avoid biases, hyperparameter tuning to improve the reliability of models, and estimating proper confidence intervals around inferences using a bootstrap sampling with replacement strategy. We computationally confirm the benefits of the proposed ML workflow by using a simulated dataset with nonlinear and non-monotonic exposure-response relationships.

Uncertainty quantification of predicting stable structures for high-entropy alloys using Bayesian neural networks

Uncertainty quantification of predicting stable structures for high-entropy alloys using Bayesian neural networks

Label-Free Fault Detection Scheme for Inverters of PV Systems: Deep Reinforcement Learning-Based Dynamic Threshold

Generally, photovoltaic (PV) fault detection approaches can be divided into two groups: end-to-end and threshold methods. The end-to-end method typically uses a deep neural network (DNN) to learn fault patterns from labeled datasets, which directly detect whether faults occur or not. The threshold method first estimates power generation and uses thresholds to detect atypical deviations of measured values from estimated ones. The former method heavily relies on fault-labeled data and, therefore, requires the collection of abnormal event records, which is usually difficult, due to the sparseness of these events. The latter method typically uses statistical approaches, such as 3-sigma, to find thresholds, and it can be practically utilized without fault labels. However, setting a threshold with a proper confidence interval is still challenging, as PV power generation is sensitive to variations in environmental conditions, such as irradiance, ambient temperature, wind speed and humidity. In this paper, we propose a novel deep reinforcement learning (DRL)-based label-free fault detection scheme in which thresholds are dynamically assigned with suitable confidence intervals under varying environmental conditions. Various weather properties were used as input features (i.e., states) to a DRL agent, and proper thresholds were estimated in real time from the actions of the DRL agent. To this end, a reward function was designed for learning proper thresholds without fault labels under different weather conditions. To evaluate the performance of the proposed scheme, the PV dataset of the National Institute of Standards and Technology (NIST) was used, as it includes paired records of local weather and PV generations. The DRL-based scheme was compared with static and conventional dynamic threshold methods, based on statistical approaches. The results revealed that the proposed scheme outperformed the existing methods, providing a 5.67% higher F1-score in the NIST dataset.

Confidence Indexing of Automated Detected Synsets: A Case Study on Contemporary Turkish Dictionary

In this study, a novel confidence indexing algorithm is proposed to minimize human labor in controlling the reliability of automatically extracted synsets from a non-machine-readable monolingual dictionary. Contemporary Turkish Dictionary of Turkish Language Association is used as the monolingual dictionary data. First, the synonym relations are extracted by traditional text processing methods from dictionary definitions and a graph is prepared in Lemma-Sense network architecture. After each synonym relation is labeled by a proper confidence index, synonym pairs with desired confidence indexes are analyzed to detect synsets with a spanning tree-based method. This approach can label synsets with one of three cumulative confidence levels (CL-1, CL-2, and CL-3). According to the confidence levels, synsets are compared with KeNet which is the only open access Turkish Wordnet. Consequently, while most matches with the synsets of KeNet is determined in CL-1 and CL-2 confidence levels, the synsets determined at CL-3 level reveal errors in the dictionary definitions. This novel approach does not find only the reliability of automatically detected synsets, but it can also point out errors of detected synsets from the dictionary.

How to Prepare for the Membership of the Royal College of Physicians (UK) Part 2 Clinical Examination (Practical Assessment of Clinical Examination Skills): An Exploratory Study

The Practical Assessment of Clinical Examination Skills (PACES) examination is considered the most difficult component of the three-part Membership of the Royal College of Physicians (MRCP) United Kingdom (UK) diploma, which provides budding physicians with access to a diverse pathway of subspecialties and a rewarding medical career. Preparing for the exam is onerous, especially when also juggling work and family demands. This research explored how candidates can prepare for this exam in an organised manner to improve their chances of first-pass success. Semi-structured interviews were conducted with 10 participants in the West Yorkshire region in the UK who had passed the PACES examination within 24 months. The interviews were transcribed onto QSR NVivo software, coded reflexively and analysed using thematic analysis. The analysis identified 25 codes that were aggregated into the following themes according to their density; (a) mental conditioning, (b) clinical exposure, (c) external factors and support, (d) revision courses, (e) knowledge acquisition, (f) playing the objective structured clinical examination (OSCE) game, (g) change of strategy, and (h) novel methods. In summary, it takes more than just clinical exposure and rote learning to pass the exam; preparation involves a complex, multifaceted approach. With proper mental conditioning, organisational skills and confidence, the adverse effect of test anxiety could be reduced to improve chances of success. Peer learning is an effective tool for facilitating examination preparation among candidates, besides providing a supportive framework. Cues from this study could act as a guiding framework for future PACES candidates to further refine their study strategies.

Timely Combination Therapy of Skeletal and Dental Class II malocclusion with Temporary Anchorage Device and Mesially Impacted Mandibular 2ndMolars: A Case-Report

Introduction: Patient is a 14 years old Chinese boy presented with skeletal Class II with incompetent lips, severe dental Class II malocclusion and severe crowding complicated with both sides mesio-angular impacted of mandibular 2nd molars.
 Treatment Plan: Growth modification using functional appliance for skeletal Class II correction with maxillary arch expansion and correction of mandibular plane angle by intrusion of maxillary molars. Extraction of lower 1st premolars and 2nd molars on both sides was carried out in conjunction with fixed orthodontic alignment and to facilitate of spontaneous eruption of lower third molar. Midline correction was followed by space closure and finishing and detailing.
 Results: Alignment of upper arch in the regained space with maxillary expansion without any extraction was done. Anterior maxillary dentitions were retracted using Temporary Skeletal Anchorage Devices (TSADs). In lower arch 1st molars were mesially protracted by using class II elastics. 3rd molars spontaneously erupted in the extraction space of 2nd molar. Midline corrected with good facial profile and competent lips.
 Conclusions: Timely execution of expert treatment plan can bring cost effective and time saving results with minimum loss. By proper counseling, patient compliance and confidence can be boosted which could have a positive effect on the treatment outcome.
 Bangladesh Journal of Medical Science Vol.20(3) 2021 p.682-690

The special demands on airborne geophysics of engineering projects

Airborne geophysics can be very useful in many engineering projects, but geophysicists need to learn to understand and address the needs of the engineering problem. Geophysicists have to design surveys and predict results for never-been-done applications with very little prior information or experience. It is essential to educate the engineers on what geophysics can do and what the limits are, show how geophysics should be properly integrated into the project. Finally, geophysicists must provide data and interpretation in formats and measurements most useful to the project, with proper confidence factors provided for the interpreted results.

Detecting SARS-CoV-2 From Chest X-Ray Using Artificial Intelligence.

Chest radiographs (X-rays) combined with Deep Convolutional Neural Network (CNN) methods have been demonstrated to detect and diagnose the onset of COVID-19, the disease caused by the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). However, questions remain regarding the accuracy of those methods as they are often challenged by limited datasets, performance legitimacy on imbalanced data, and have their results typically reported without proper confidence intervals. Considering the opportunity to address these issues, in this study, we propose and test six modified deep learning models, including VGG16, InceptionResNetV2, ResNet50, MobileNetV2, ResNet101, and VGG19 to detect SARS-CoV-2 infection from chest X-ray images. Results are evaluated in terms of accuracy, precision, recall, and f- score using a small and balanced dataset (Study One), and a larger and imbalanced dataset (Study Two). With 95% confidence interval, VGG16 and MobileNetV2 show that, on both datasets, the model could identify patients with COVID-19 symptoms with an accuracy of up to 100%. We also present a pilot test of VGG16 models on a multi-class dataset, showing promising results by achieving 91% accuracy in detecting COVID-19, normal, and Pneumonia patients. Furthermore, we demonstrated that poorly performing models in Study One (ResNet50 and ResNet101) had their accuracy rise from 70% to 93% once trained with the comparatively larger dataset of Study Two. Still, models like InceptionResNetV2 and VGG19’s demonstrated an accuracy of 97% on both datasets, which posits the effectiveness of our proposed methods, ultimately presenting a reasonable and accessible alternative to identify patients with COVID-19.

Self-adaptive shared control with brain state evaluation network for human-wheelchair cooperation

Objective. For the shared control systems, how to trade off the control weight between robot autonomy and human operator is an important issue, especially for BCI-based systems. However, most of existing shared controllers have paid less attention to the effects caused by subjects with different levels of brain control ability. Approach. In this paper, a brain state evaluation network, termed BSE-NET, is proposed to evaluate subjects’ brain control ability online based on quantized attention-gated kernel reinforcement learning. With the output of BSE-NET (confidence score), a shared controller is designed to dynamically adjust the control weight between robot autonomy and human operator. Main results. The experimental results show that most of subjects achieved high and stable experimental success rate of approximately 90%. Furthermore, for subjects with different accuracy on EEG decoding, a proper confidence score can be dynamically generated to reflect their levels of brain control ability, and the proposed system can effectively adjust the control weight in all-time shared control. Significance. We discuss how our proposed method shows promise for BCI applications that can evaluate subjects’ brain control ability online as well as provide a method for the research on self-adaptive shared control to adaptively balance control weight between subject’s instruction and robot autonomy.

Statistical methods in sterile neutrino experiments

We compare the current statistical methods applied to short-baseline neutrino oscillation experiments looking for light sterile neutrinos. The interpretation of the reconstructed confidence regions differs because of the different definition of the alternative hypothesis in the hypothesis test. When the alternative hypothesis extends over the whole physical allowed parameter space of the oscillation parameters, proper confidence regions are obtained both in case of a limit and a signal. Since the different definition of the alternative hypothesis prevents from a direct comparison of the results, a standardized analysis is proposed.

PEDAGOGICAL PRACTICE AS A LINK FOR CONNECTING THEORY AND PRACTICE IN TEACHING ELEMENTARY MATHEMATICS

Mathematics as a subject has been taught since the early years of the education. Elementary school teachers take great responsibility to teach this difficult, abstract, but sometimes boring subject. Hypothetically privilege of understanding this subject is a certain number of students. Teachers not only need to be a professional of the subject, but at the same time they must be artists also. This requires much work, patience, determination, and much love for children and the profession. Students of the Pedagogy Faculty of University of Tetovo are getting prepared for teaching Mathematics in elementary schoolsfirst in the theoretical part and then in the practical part. Pedagogical practice is realized in elementary schools, observing Mathematic teaching by teachers with experience and realizations of Mathematics classes by students themselves. Mathematical observations enable students to benefit from the experience of other teachers. This observation develops students observation skills, develops communication skills with colleagues of the same profession. Observation also makes it possible for students to observe application in practice of knowledge in the pedagogical content. This research evaluates the impact of Mathematic teaching practice on Pedagogy Faculty students of Tetova University. Questionnaire collects data of 46 students. Results showes that PedagoyFacullty students benefited greatly from participating in Mathematics teaching practice because they were able to build proper confidence and competence in lesson preparation and developed skills and attitude of a teacher during the practice. It was also revealed that teaching practice helps students to learn how to participate in school activities. This study proves (Pearson coefficient) the hypothesis that students’ opinion is not depended by students general success and students’ Mathematics success. According the hypothesis Mathematics teaching practice benefits all students and it should be practiced without hesitation in the subject Didactics of Teaching Mathematics as well as in other subjects where didactic theories are studied.

Significance Tests for Boosted Location and Scale Models with Linear Base-Learners.

Generalized additive models for location scale and shape (GAMLSS) offer very flexible solutions to a wide range of statistical analysis problems, but can be challenging in terms of proper model specification. This complex task can be simplified using regularization techniques such as gradient boosting algorithms, but the estimates derived from such models are shrunken towards zero and it is consequently not straightforward to calculate proper confidence intervals or test statistics. In this article, we propose two strategies to obtain p-values for linear effect estimates for Gaussian location and scale models based on permutation tests and a parametric bootstrap approach. These procedures can provide a solution for one of the remaining problems in the application of gradient boosting algorithms for distributional regression in biostatistical data analyses. Results from extensive simulations indicate that in low-dimensional data both suggested approaches are able to hold the type-I error threshold and provide reasonable test power comparable to the Wald-type test for maximum likelihood inference. In high-dimensional data, when gradient boosting is the only feasible inference for this model class, the power decreases but the type-I error is still under control. In addition, we demonstrate the application of both tests in an epidemiological study to analyse the impact of physical exercise on both average and the stability of the lung function of elderly people in Germany.

Uncertainty quantification and propagation in CALPHAD modeling

Design is about making decisions bounded by a quantifiable degree of certainty. In the context of alloy design, Integrated Computational Materials Engineering (ICME) provides the framework whereby performance requirements are ultimately transformed into alloy/processing specifications through the combination of (complex) computational models connecting process-structure-property-performance relationships and experiments. Most ICME approaches consider the models used as deterministic and thus do not provide the means to make alloy design decisions with proper confidence measures. At the root of ICME lie CALPHAD models that describe the thermodynamics and phase stability of phases under specific thermodynamic boundary conditions. To date, the vast majority of efforts within the CALPHAD community have been deterministic in that thermodynamic models and the resulting thermodynamic properties and phase diagram features do not explicitly account for the uncertainties inherent in the model formulation or in the experimental/computational data used. In this contribution, we provide an overview of the state of the field. We review major efforts thus far and we then provide a (brief) tutorial on basic concepts of uncertainty quantification and propagation (UQ/UP) in CALPHAD. We discuss the major features of frequentist and Bayesian interpretations of uncertainty and proceed with a discussion of recent case studies in which UQ has been used to parameterize models for the thermodynamic properties of phases. We follow our discussion by presenting frameworks and demonstrating the propagation of uncertainty in thermodynamic properties and phase diagram predictions and briefly discuss how we can use Bayesian frameworks for rigorous model selection as well as for model fusion. We close our contribution by providing context for what has been done and what remains to be accomplished in order to fully embrace the management of uncertainty in CALPHAD modeling, a foundational element of ICME.