Confidence Interval Estimation Research Articles

S936 Factors Associated With Fecal Calprotectin Sample Collection Compliance: An IBD Center Quality Improvement Project

Introduction: Regular assessment of objective markers of inflammatory bowel disease (IBD) activity is recommended. Fecal calprotectin (Fcal) is a non-invasive and inexpensive biomarker of disease activity, however, patient compliance with this test is variable and incompletely described. We assessed compliance rates with Fcal tests and identified factors associated with non-compliance. Methods: As a quality improvement project, we conducted a retrospective chart review of IBD patients who had a Fcal test ordered by us between August 2021 and December 2021. Demographic, clinical, disease-related, and test-related (location of lab relative to the patient’s visit in clinic or at home) information were recorded to determine predictors for non-compliance. Simple statistical analysis and multivariable logistic regression was performed. Results: Of 303 patients, 165 (54.4%) had an order for Fcal. Of the Fcal tests ordered, 55 (33.3%) were not completed. Between those with complete versus incomplete tests, IBD remission status (67.8% vs. 83.7%, p = 0.033), history of prior Fcal order completion (93.2% vs. 68.4%, p = 0.001), and test ordered to an external site (62.7% vs. 85.5%, p = 0.004) were significantly different. A multivariable logistic regression with these factors as well as age, gender, telehealth vs. in-person consultation, and patient residence revealed that history of a prior completed Fcal test is independently associated with subsequent test completion (odds ratio (OR) = 8.3, 95% CI 1.9-35.5, p = 0.004). Fcal testing location (institutional or third-party) was marginally significant (OR = 0.27, 95% CI 0.7-1.0, p = 0.051). A multivariable linear regression for days between test order and completion found that tests ordered for external locations are associated with delayed completion (95% CI 5.2-20.6, p = 0.002). Conclusion: In this single center analysis with Fcal testing in patients with IBD, we found that a history of incomplete testing is associated with subsequent non-compliance, and distant location of the lab is associated with delayed completion of the test. These findings suggest that ordering providers should identify additional methods to educate on the importance of disease monitoring with Fcal and when possible, simplify the lab used for testing. Future availability of at home Fcal tests may be helpful in this regard. Table 1. - Linear Regressions for Predictors of Fecal Calprotectin Compliance Logistic Regression for Test Compliance Regression for Delayed Testing Completion Estimate Effect Size (B) Standard Error P - Value 95% Confidence Interval Estimate Effect Size (B) Standard Error P - Value 95% Confidence Interval Age 0.017 0.015 0.248 0.998-1.048 0.148 0.104 0.163 -31.309-7.575 Gender -0.441 0.543 0.417 0.222-1.866 0.877 3.635 0.810 0.062-0.357 IBD Remission -0.949 0.624 0.128 0.114-1.316 -0.331 3.895 0.933 -6.421-8.176 Consultation Type -.0523 0.528 0.322 0.210-1.570 0.102 3.992 0.980 -8.151-7.490 History of Fecal Calprotectin Completion 2.116 0.741 0.004 1.942-35.493 6.628 2.625 0.357 -7.699-20.956 Fecal Calprotectin Testing Location -1.304 0.668 0.51 0.073-1.005 12.875 3.843 0.002 5.160-20.591

A Hotel Ranking Model Through Online Reviews With Aspect-Based Sentiment Analysis

The number of online textual reviews on each hotel aspect can reflect the tourist preference difference on distinct aspects. Therefore, not only online textual reviews but their numbers have a significant impact on tourists’ hotel selection decisions. Motivated by this observation, this study proposes a hotel ranking model for hotel selection based on the sentiment analysis of online textual reviews by considering the differences in the number of reviews on different aspects. We explicitly model the differences in the number of reviews on aspects through the confidence interval estimation. In addition, the AS-Capsules model, which can jointly perform aspect detection and aspect-level sentiment classification with high accuracy, is employed for sentiment analysis. We conducted a case study on TripAdvisor.com, the experimental results show that our proposed model is able to effectively assist the tourists in making the desirable decision on hotel selection.

Key Interpretation Challenges for Wastewater-Based Epidemiology of Illicit Drugs: A Norwegian Three-City Case Study

Introduction: Wastewater-based epidemiology (WBE) has emerged as a timely, non-invasive, and cost-effective indicator of illicit drug consumption. It is increasingly used by international organizations as a proxy measure for estimates of drug prevalence and related trends. Nevertheless, the literature exploring the limitations of WBE remains limited. This paper aims to shed further light on important shortcomings of WBE with recommendations on moving forward. Method: Utilizing case study and statistical analysis, the paper critically reviews methodological challenges associated with WBE results related to (i) levels, (ii) trends, and (iii) between-city comparisons of drug use. Data from raw influent wastewater samples from wastewater plants in the cities of Oslo, Bergen and Stavanger/Sandnes were analysed for amphetamine, methamphetamine, MDMA, and cocaine (benzoylecgonine) over a 3-year period. Normalized population loads were calculated and variation in daily loads analysed with plots and estimation of means, confidence intervals, and coefficient of variation. Linear regression models examined trends and between-city differences. Results: Plots and statistical analyses revealed extensive variation in daily loads, with min/max values of 6.1/453.9 mg/day per 1,000 inhabitants 15–64 years for amphetamine and correspondingly 9.4/675.9 mg for methamphetamine. Substantial differences in load levels and patterns across time and plants were also observed. A carefully designed sampling procedure and a relatively large number of daily samples are required to obtain estimates of sufficient precision for determining trends in space or time. Cross-referencing with alternative trend variables can improve the interpretation of WBE trend indicators. Finally, when using mean load levels for different wastewater-treatment plants to assess spatial variation in drug use, the representativeness of the catchment area should be evaluated before interpreting observed changes as city differences. Conclusion: Although WBE is a useful supplementary indicator of illicit drug consumption, important methodological issues and potential shortcomings should be taken into account when designing sampling procedures and interpreting the analytical results.

Research on multiaxial fatigue life of notched specimens based on Weibull distribution and Bayes estimation

In this study, based on the nominal stress method and the influence of stress gradient on fatigue life, a fatigue life prediction method combined with the Weibull distribution is proposed. Firstly, the strain energy density is defined as the damage parameter to determine the location of the critical plane. Then, the equivalent stress function is established by extracting the equivalent stress of the specific path on the critical plane, which is used to establish the modified nominal stress method by combining with the stress gradient factor. Secondly, the Weibull distribution is combined to establish a new fatigue life prediction model of the notched specimens under different failure probabilities. The accuracy of the model is verified with three materials, then the prediction life of the proposed model is compared with that of the SWT model, the FS model and the Manson-Coffin equation. The results show that the accuracy of the proposed model is higher than the other three models. Finally, the reliability of the proposed model is evaluated by the lower confidence limit (LCL) estimation and interval estimation to explore the feasibility of this method. In addition, the P-S-N curves under different failure probabilities are established to guide actual projects.

Use of low molecular weight heparin and hemoglobin fall in COVID-19 patients: A STROBE-compliant study

In patients with coronavirus disease 2019 (COVID-19), anticoagulation was suggested as a mitigating strategy. However, little research has been conducted on the adverse consequences of anticoagulant medication. This study aimed to investigate the adverse effect of low molecular weight heparin (LMWH) on hemoglobin fall in COVID-19 treatment. The electronic medical records of COVID-19 patients with pneumonia were collected (including clinical characteristics, vaccination status, complete blood count, coagulation profile, inflammatory cytokines, serum biochemical indicators, and computerized tomography imaging score). Whether they received LMWH, patients were divided into the LMWH group and the control group. Count data were represented as frequency distribution, and a 2-tailed test was used to compare the 2 groups. Spearman rank correlation was used to evaluate the interrelation between changes in hemoglobin and LMWH. The confounding factors were excluded by logistic regression analysis. A total of 179 COVID-19 pneumonia patients were enrolled (81 in the LMWH group and 98 in the control group). The change in hemoglobin was −6.0g/L (IQR −10.8 to 1.0) in the LMWH group and −2.0g/L (IQR −7.0 to 4.0) in the control group (P < .001, between-group difference, −5.0 g/L; 95% confidence interval, −7.0 to −3.0, calculated with the use of the Mann–Whitney U test and the Hodges–Lehmann estimate of confidence intervals for pseudo-medians). The results of multivariate regression analysis showed that after adjusting for confounding factors, LMWH use was not associated with a decrease in hemoglobin (P > .05). In nonsevere COVID-19 patients with pneumonia, the preventive use of LMWH did not lower hemoglobin.

Confidence interval estimation for fingerprint-based indoor localization

Fingerprint-based localization methods provide high accuracy location estimation, which use machine learning algorithms to recognize the statistical patterns of collected data. In these methods, the users’ locations can be estimated based on the received signal strength vectors from some transmitters. However, the data collection is a labor-intensive phase, and the collected data should be updated periodically. Many researchers have contributed to reducing this cost. The easiest way to remove the data collection cost is to use fingerprints generated by the model-based approaches, in which the trained machine learning algorithm can be updated based on the environment changes. Probabilistic-based localization algorithms, in addition to the user location, can estimate a region of interest called 2σ confidence interval in which the probability of user presence is 95%. Gaussian process regression (GPR) is a probabilistic method that can be used to achieve this goal. However, conventional GPR (CGPR) cannot accurately estimate the confidence interval when noise-free fingerprints generated by the model-based approaches are used in the training phase. In this paper, we propose a novel GPR-based localization algorithm, named enhanced GPR (EGPR), which improves the accuracy level of confidence interval estimation compared to the existing methods while fixing the level of computational complexity in the online phase. We also theoretically prove that GPR-based algorithms are minimum variance unbiased and efficient estimators. Experiments under line-of-sight and non-line-of-sight conditions demonstrate the superiority of our proposed method over counterparts in terms of accuracy as well as applicability in real-time localization systems.

Bayesian Approach for Confidence Intervals of Variance on the Normal Distribution

This research aims to compare estimating the confidence intervals of variance based on the normal distribution with the primary method and the Bayesian approach. The maximum likelihood is the well-known method to approximate variance, and the Chi-squared distribution performs the confidence interval. The central Bayesian approach forms the posterior distribution that makes the variance estimator, which depends on the probability and prior distributions. Most introductory prior information looks for the availability of the prior distribution, informative prior distribution, and noninformative prior distribution. The gamma, Chi-squared, and exponential distributions are defined in the prior distribution. The informative prior distribution uses the Markov Chain Monte Carlo (MCMC) method to draw the random sample from the posterior distribution. The Fisher information performs the Wald confidence interval as the noninformative prior distribution. The interval estimation of the Bayesian approach is obtained from the central limit theorem. The performance of these methods considers the coverage probability and minimum value of the average width. The Monte Carlo process simulates the data from a normal distribution with the true parameter of mean and several variances and the sample sizes. The R program generates the simulated data repeated 10,000 times in each situation. The results showed that the maximum likelihood method employed on the small sample sizes. The best confidence interval estimation was when sample sizes increased the Bayesian approach with an available prior distribution. Overall, the Wald confidence interval tended to outperform the large sample sizes. For application in real data, we expressed the reported airborne particulate matter of 2.5 in Bangkok, Thailand. We used the 10–1000 records to estimate the confidence interval of variance and evaluated the interval width. The results are similar to those of the simulation study.

Sublingual Dexmedetomidine for Agitation Associated with Schizophrenia or Bipolar Disorder: A Post Hoc Analysis of Number Needed to Treat, Number Needed to Harm, and Likelihood to be Helped or Harmed.

IntroductionThe objective was to use the evidence-based medicine metrics of number needed to treat, number needed to harm, and likelihood to be helped or harmed to appraise the clinical efficacy and tolerability of sublingual dexmedetomidine in adults with agitation associated with schizophrenia or bipolar disorder.MethodsSublingual dexmedetomidine data for this post hoc analysis were obtained from two similarly designed, double-blind, randomized, placebo-controlled studies of adults with schizophrenia or bipolar disorder. Response to treatment was defined as a ≥ 40% reduction from baseline in the Positive and Negative Syndrome Scale-Excited Component (PEC). Tolerability was assessed by evaluating rates of adverse events.ResultsThe number needed to treat (95% confidence interval) estimate versus placebo for PEC response at 2 h post-dose was 3 (2, 3) for the sublingual dexmedetomidine 180-µg group (n = 125) and 3 (3, 4) for the 120-µg group (n = 129) in the study of patients with schizophrenia and 3 (2, 3) for the sublingual dexmedetomidine 180-µg group (n = 126) and 4 (3, 6) for the 120-µg group (n = 126) in the study of patients with bipolar disorder. Number needed to harm values versus placebo were greater than 10 for all adverse events except somnolence, where the number needed to harm (95% confidence interval) was 7 (5, 10) for all doses pooled from both studies. In all instances, likelihood to be helped or harmed values were greater than 1 for efficacy versus applicable tolerability outcomes.ConclusionsThe number needed to treat, number needed to harm, and likelihood to be helped or harmed of sublingual dexmedetomidine support a favorable benefit–risk profile in adults with acute agitation associated with schizophrenia or bipolar disorder.Trial RegistrationClinicalTrials.gov, https://clinicaltrials.gov/ct2/show/NCT04268303, NCT04268303. ClinicalTrials.gov, https://clinicaltrials.gov/ct2/show/NCT04276883, NCT04276883.Supplementary InformationThe online version contains supplementary material available at 10.1007/s12325-022-02274-3.

Confidence interval estimation for the difference and ratio of the means of two gamma distributions

Gamma distribution is widely used in applied fields due to its flexibility of accommodating right-skewed data. This article explores statistical methods for constructing confidence intervals for both the difference and ratio of two gamma means. We propose several methods based on the concepts of generalized inference, variance estimates recovery (MOVER) and parametric bootstrapping. The performances of proposed methods are evaluated and compared via a comprehensive simulation study. Simulation results show that the hybrid method that combines MOVER with parametric bootstrapping can provide confidence intervals with reasonable coverage probabilities and interval lengths even for parameter settings with small shape parameters. Finally, two real data examples from environmental and engineering studies are analyzed using the proposed methods.

N200 and P300 component changes in Parkinson's disease: a meta-analysis.

Cognitive impairment can seriously affect the quality of life of Parkinson's disease (PD) patients. Although numerous studies showed that N200, P300 latency and amplitude are correlated with cognitive functions, there is a sufficient amount of controversial results. Therefore, it is necessary to conduct a meta-analysis of N200, P300 latency and amplitude data of event-related potential (ERP) in PD. We systematically searched on PubMed and Web of Science for PD-related ERP studies published before December 2021. Standard mean difference (SMD) and 95% confidence interval (CI) estimates of N200 and P300 components were compared among PD patients, PD dementia (PDD) patients, PD non-dementia (PDND) patient, and healthy control (HC). Our meta-analysis showed prolonged N200 latency at the Fz, Cz electrode sites, prolonged P300 latency at the Fz sites in PD patients, compared to HC; prolonged N200 latency at the Cz, Pz electrode sites in PDND patients, compared to HC; prolonged P300 latency at the Cz site in PDD patients, compared to PDND patients; and reduced P300 amplitude at the Fz electrode site in PDND patients, compared to HC. N200 and P300 component may be potential electrophysiological biomarkers of early cognitive impairment in PD patients. Future studies are needed to confirm this conclusion. Estimates of N200 and P300 component can be a valuable support for clinicians in diagnosis of early cognitive impairment in PD patients due to the simplicity and non-invasiveness of the procedure.

Confidence interval estimation for vaccine efficacy against COVID-19.

This article focuses on the construction of a confidence interval for vaccine efficacy against contagious coronavirus disease-2019 (COVID-19) in a fixed number of events design. Five different approaches are presented, and their performance is investigated in terms of the two-sided coverage probability, non-coverage probability at the lower tail, and expected confidence interval width. Furthermore, the effect of under-sensitivity of diagnosis tests on vaccine efficacy estimation was evaluated. Except for the exact conditional method, the non-coverage probability of the remaining methods may exceed the nominal significance level, e.g., 5%, even for a large number of total confirmed COVID-19 cases. The narrower confidence interval width from the Bayesian, approximate Poisson, and mid-P methods are on the cost of increased instability of coverage probability. When the sensitivity of diagnosis test in the vaccine group is lower than that in the placebo group, the reported vaccine efficacy tends to be overly optimistic. The exact conditional method is preferable to other methods in COVID-19 vaccine efficacy trials when the total number of cases reaches 60; otherwise, mid-p method can be used to obtain a narrower interval width.

An optimal purely sequential strategy with asymptotic second-order properties: Applications from statistical inference and data analysis

We develop a new class of purely sequential methodologies under an assumption that the population distribution belongs to a location-scale family. Both asymptotic first-order and second-order theories are put forward with substantial generality under a big and unified tent that successfully lead to a broad set of illustrations. After we identify an appropriately defined optimal strategy under this unified structure, we introduce applications that handle a variety of interesting inference problems. These are associated with, but not limited to, the following areas: (a) the fixed-width confidence interval (FWCI) estimation, (b) the minimum risk point estimation (MRPE), (c) the fixed-size confidence region (FSCR) estimation, (d) multiple comparisons, and (e) selecting the best normal treatment (StBNT). In illustrations (a)–(d), we have highlighted a number of choices of population distributions. Some illustrations are accompanied with data analyses.

Real-time fully automated dosimetric computation for CT images in the clinical workflow: A feasibility study.

BackgroundCurrently, the volume computed tomography dose index (CTDIvol), the most-used quantity to express the output dose of a computed tomography (CT) patient’s dose, is not related to the real size and attenuation properties of each patient. The size-specific dose estimates (SSDE), based on the water-equivalent diameter (D W) overcome those issues. The proposed methods found in the literature do not allow real-time computation of D W and SSDE.PurposeThis study aims to develop a software to compute D W and SSDE in a real-time clinical workflow.MethodIn total, 430 CT studies and scans of a water-filled funnel phantom were used to compute accuracy and evaluate the times required to compute the D W and SSDE. Two one-sided tests (TOST) equivalence test, Bland–Altman analysis, and bootstrap-based confidence interval estimations were used to evaluate the differences between actual diameter and D W computed automatically and between D W computed automatically and manually.ResultsThe mean difference between the D W computed automatically and the actual water diameter for each slice is −0.027% with a TOST confidence interval equal to [−0.087%, 0.033%]. Bland–Altman bias is −0.009% [−0.016%, −0.001%] with lower limits of agreement (LoA) equal to −0.0010 [−0.094%, −0.068%] and upper LoA equal to 0.064% [0.051%, 0.077%]. The mean difference between D W computed automatically and manually is −0.014% with a TOST confidence interval equal to [−0.056%, 0.028%] on phantom and 0.41% with a TOST confidence interval equal to [0.358%, 0.462%] on real patients. The mean time to process a single image is 13.99 ms [13.69 ms, 14.30 ms], and the mean time to process an entire study is 11.5 s [10.62 s, 12.63 s].ConclusionThe system shows that it is possible to have highly accurate D W and SSDE in almost real-time without affecting the clinical workflow of CT examinations.

Risk factors associated with unplanned ICU admissions following paediatric surgery: A systematic review.

Background Unplanned admissions to the intensive care unit (ICU) have important implications in the general management of patients. Research in this area has been conducted in the adult and non-surgical population. To date, there is no systematic review addressing risk factors in the paediatric surgical population. Objectives To synthesise the information from studies that explore the risk factors associated with unplanned ICU admissions following surgery in children through a systematic review process. Methods We conducted a systematic review of published literature (PROSPERO registration CRD42020163766), adhering to the Preferred Reporting of Observational Studies and Meta-Analysis (PRISMA) statement. The Population, Exposure, Comparator, Outcome (PECO) strategy used was based on: population – paediatric population, exposure – risk factors, comparator – other, and outcome – unplanned ICU admission. Data that reported on unplanned ICU admissions following paediatric surgery were extracted and analysed. Quality of the studies was assessed using the Newcastle-Ottawa Scale. Results Seven studies were included in the data synthesis. Four studies were of good quality with the Newcastle-Ottawa Scale score ≥7 points. The pooled prevalence (95% confidence interval) estimate of unplanned ICU stay was 2.69% (0.05 - 8.6%) and ranged between 0.06% and 8.3%. Significant risk factors included abnormal sleep studies and the presence of comorbidities in adenotonsillectomy surgery. In the general surgical population, younger age, comorbidities and general anaesthesia were significant. Abdominal surgery and ear, nose and throat (ENT) surgery resulted in a higher risk of unplanned ICU admission. Owing to the heterogeneity of the data, a meta-analysis with risk prediction could not be performed. Conclusion Significant patient, surgical and anaesthetic risk factors associated with unplanned ICU admission in children following surgery are described in this systematic review. A combination of these factors may direct planning toward anticipation of the need for a higher level of postoperative care. Further work to develop a predictive score for unplanned ICU stay is desirable. Contributions of the study Unplanned admissions to the intensive care unit (ICU) have been acknowledged as an overall marker of safety.[1] Awareness of this concept has encouraged research to determine the incidence and risk factors of these occurrences. This research has been interrogated in a systematic review process with beneficial conclusions drawn; however, these studies included adults and non-surgical patients.[2–4] To date, we have not been able to find a systematic review addressing the risk factors associated with unplanned ICU admissions in paediatric surgical patients.

Sampling plan for socioeconomic development indicators in Brazil: practical implications when considering precision and cost

Socioeconomic development indicators such as Income, Human Development Index, and Gini Index have explained recent deforestation events in Brazil. These indicators, obtained from the demographic census, are expensive and are released every ten years. Despite being costly, they are important descriptors of the phenomenon of deforestation that occurs in one of the most relevant ecosystems in the world, the Brazilian forests. Therefore, the objective in this study was to detect the most representative and the lowest costly sample in order to obtain the referred indicators, from Brazilian municipalities from 2000 to 2010. The sample mean and confidence interval were obtained to be compared with the population average. The mean and confidence interval estimators come from 1,000 resamples according to simple random sampling and stratified random by geographic region. We evaluated the following sample intensities: 10%, 20%, 30%, 40% and 50% of the Brazilian municipalities. The results indicated that: a) 10% of Brazilian municipalities are sufficient to represent Income, HDI and Gini; b) We recommend both random and stratified sampling when selecting the municipalities that will provide the HDI and Gini values, on the contrary, for Income, the indicated sampling is the stratified one. These results show the importance of applied research in sampling and simulation to optimize the survey of demographic data collection, in a situation of budgetary constraint, in countries of vast dimensions such as Brazil. This study can increase understanding of the importance of socioeconomic development indicators to estimating progressive forest loss in Brazil, the country with the largest rainforest in the world.

Assessments of epistemic uncertainty using Gaussian stochastic weight averaging for fluid-flow regression

We use Gaussian stochastic weight averaging (SWAG) to assess the epistemic uncertainty associated with neural-network-based function approximation relevant to fluid flows. SWAG approximates a posterior Gaussian distribution of each weight, given training data, and a constant learning rate. Having access to this distribution, it is able to create multiple models with various combinations of sampled weights, which can be used to obtain ensemble predictions. The average of such an ensemble can be regarded as the ‘mean estimation’, whereas its standard deviation can be used to construct ‘confidence intervals’, which enable us to perform uncertainty quantification (UQ) with regard to the training process of neural networks. We utilize representative neural-network-based function approximation tasks for the following cases: (i) a two-dimensional circular-cylinder wake; (ii) the DayMET dataset (maximum daily temperature in North America); (iii) a three-dimensional square-cylinder wake; and (iv) urban flow, to assess the generalizability of the present idea for a wide range of complex datasets. SWAG-based UQ can be applied regardless of the network architecture, and therefore, we demonstrate the applicability of the method for two types of neural networks: (i) global field reconstruction from sparse sensors by combining convolutional neural network (CNN) and multi-layer perceptron (MLP); and (ii) far-field state estimation from sectional data with two-dimensional CNN. We find that SWAG can obtain physically-interpretable confidence-interval estimates from the perspective of epistemic uncertainty. This capability supports its use for a wide range of problems in science and engineering.

Inferences of the Multicomponent Stress–Strength Reliability for Burr XII Distributions

Multicomponent stress–strength reliability (MSR) is explored for the system with Burr XII distributed components under Type-II censoring. When the distributions of strength and stress variables have Burr XII distributions with common or unequal inner shape parameters, the existence and uniqueness of the maximum likelihood estimators are investigated and established. The associated approximate confidence intervals are obtained by using the asymptotic normal distribution theory along with the delta method and parametric bootstrap procedure, respectively. Moreover, alternative generalized pivotal quantities-based point and confidence interval estimators are developed. Additionally, a likelihood ratio test is presented to diagnose the equivalence of both inner shape parameters or not. Conclusively, Monte Carlo simulations and real data analysis are conducted for illustration.

An efficient confidence interval estimation for prevalence calculated from misclassified data

The estimation of prevalence using a screening tool is done frequently in epidemiology research. The tools used for the estimation are usually associated with a certain level of misclassification. Additional adjustments are required to eliminate the bias in the prevalence and the confidence interval (CI) estimate. A frequently used method for this correction is by modifying the upper and lower limits, using sensitivity and specificity, increasing the width of the CI. The issue is exaggerated with a minimal sample size. Zhou and Li recently developed a method to estimate the CI using the Edgeworth expansion of the logit transformed binomial proportion. This article introduces a specialised tool by re-estimating the confidence limits adjusting for misclassified measurements, and assesses their characteristics through a simulation. The paper provides evidence that the re-estimated new interval performs better in the presence of misclassification.

Subsampling bootstrap in network DEA

Data Envelopment Analysis (DEA), provides an empirical estimation of the production frontier, based on an observed sample of decision making units (DMUs). Except for the single input-single output case, the asymptotic distribution of the DEA estimator can only be approximated through bootstrapping approaches. Therefore, bootstrapping techniques have been widely applied in the DEA literature to make statistical inference for the cases when the production process has a single-stage structure. However, in many cases, the transformation of inputs into outputs has an inner structure that needs to be considered. This paper examines the applicability of the subsampling bootstrap procedure in the approximation of the asymptotic distribution of the DEA estimator when the production process has a network structure, and in the presence of undesirable factors. Evidence on the performance of subsampling bootstrap is obtained through Monte Carlo experiments for the case of two-stage series structures, where overall and stage efficiency estimates are calculated using the additive decomposition approach. Results indicate great sensitivity both to the sample and subsample size, as well as to the data generating process. Subsampling methodology is then applied to construct confidence interval estimates for the overall and stage efficiency scores of railways in 22 European countries, where the railway transport process is decomposed into two stages and the railway noise pollution problem is considered as an undesirable output.

Parameters estimation approaches of the Moore and Bilikam family with reliability analysis under progressively type-II censoring

The precise investigation of the Moore and Bilikam (MB) family of distributions is essential since they cover several distributions as special cases. We evaluate different estimation procedures for parameters of MB family comprehensively. Moreover, the confidence interval and highest posterior density interval estimators of parameters are provided. The different estimation methods of the parameters of two certain cases of the MB family are compared via the simulation data. Based on the simulation approach, we realized the estimates are convergent to the real values of the parameters and the root mean squared error (RMSE) values derived by the expectation-maximization (EM) algorithm are less than Newton–Raphson (NR), stochastic EM (SEM) and Bayes estimators. Also, the higher values of the sample size lead to better estimates in the sense close to true parameter values and having smaller RMSE values. The reliability analysis of MB family is another topic in this paper, with simulation discussion. The Monte Carlo simulations are provided to assess the performances of estimation methods. We concluded the Bayesian shrinkage estimators are better than other methods of the reliability function, which have the smallest values of the RMSE. The analysis of real dataset has been presented for illustrative purposes, which confirmed simulation results.