Frequentist Setting Research Articles

Comparing frequentist and Bayesian uncertainty quantification in two‐step constitutive model calibration

AbstractThe calibration of constitutive models using experimental observations is a well‐established task in computational solid mechanics. Especially for complex constitutive models, multistep calibrations are regularly applied. While this approach is advantageous for reducing the complexity of the calibration problem, the quantification of parameter uncertainties becomes more challenging. In this study, we perform two‐step statistical parameter inference for an elasto‐plastic constitutive model utilizing a frequentist approach and Bayesian inference. The uncertainties are quantified by different methods, in particular, we compare an asymptotic normality analysis and Gaussian error propagation in the frequentist setting with sampling‐based Bayesian inference. Although frequentist and Bayesian parameter inference differ significantly in the way parameters are calibrated, both methods yield similar estimates. Additionally, the uncertainties quantified by asymptotic normality considerations and Gaussian error propagation are quite similar. However, the uncertainties of the elasticity parameters in the frequentist approach are found to be significantly smaller compared to Bayesian inference, due to the underlying approximations in the frequentist setting. Thus, uncertainties can be estimated in both settings, but it is crucial to be aware of the inherent assumptions and simplifications when interpreting the results.

Operational meanings of a generalized conditional expectation in quantum metrology

A unifying formalism of generalized conditional expectations (GCEs) for quantum mechanics has recently emerged, but its physical implications regarding the retrodiction of a quantum observable remain controversial. To address the controversy, here I offer operational meanings for a version of the GCEs in the context of quantum parameter estimation. When a quantum sensor is corrupted by decoherence, the GCE is found to relate the operator-valued optimal estimators before and after the decoherence. Furthermore, the error increase, or regret, caused by the decoherence is shown to be equal to a divergence between the two estimators. The real weak value as a special case of the GCE plays the same role in suboptimal estimation – its divergence from the optimal estimator is precisely the regret for not using the optimal measurement. For an application of the GCE, I show that it enables the use of dynamic programming for designing a controller that minimizes the estimation error. For the frequentist setting, I show that the GCE leads to a quantum Rao-Blackwell theorem, which offers significant implications for quantum metrology and thermal-light sensing in particular. These results give the GCE and the associated divergence a natural, useful, and incontrovertible role in quantum decision and control theory.

The solution surface of the Li-Stephens haplotype copying model

The Li-Stephens (LS) haplotype copying model forms the basis of a number of important statistical inference procedures in genetics. LS is a probabilistic generative model which supposes that a sampled chromosome is an imperfect mosaic of other chromosomes found in a population. In the frequentist setting which is the focus of this paper, the output of LS is a “copying path” through chromosome space. The behavior of LS depends crucially on two user-specified parameters, theta and rho , which are respectively interpreted as the rates of mutation and recombination. However, because LS is not based on a realistic model of ancestry, the precise connection between these parameters and the biological phenomena they represent is unclear. Here, we offer an alternative perspective, which considers theta and rho as tuning parameters, and seeks to understand their impact on the LS output. We derive an algorithm which, for a given dataset, efficiently partitions the (theta ,rho ) plane into regions where the output of the algorithm is constant, thereby enumerating all possible solutions to the LS model in one go. We extend this approach to the “diploid LS” model commonly used for phasing. We demonstrate the usefulness of our method by studying the effects of changing theta and rho when using LS for common bioinformatic tasks. Our findings indicate that using the conventional (i.e., population-scaled) values for theta and rho produces near optimal results for imputation, but may systematically inflate switch error in the case of phasing diploid genotypes.

P-025 Optimising stimulated IUI; a systematic review and network meta-analysis

Abstract Study question What is the effectiveness and safety of various follicular phase ovarian stimulation protocols for intrauterine insemination (IUI) performed for any indication? Summary answer Significantly higher LBR/OPR was noted for gonadotrophins versus oral agents. Clomiphene+estrogen and gonadotrophins+letrozole demonstrated significantly higher OPR/LBR compared to no stimulation/stimulation with oral agents alone. What is known already Stimulated IUI is one of the most widespread fertility treatments offered for various indications. The rationale of stimulated IUI cycles follows the concept that more available oocytes for fertilisation would lead to increase chances of conception. Although several small pairwise comparisons are reported between different stimulation agents, there is no overall comparison using a network meta-analysis. Optimising the ovarian stimulation is an important step towards evidence-based guidelines. There is a cost- effect balance which should be kept in mind between achieving multi follicular growth to boost success rates and keeping the intervention safe in relation to multiple pregnancies. Study design, size, duration This review has been conducted in accordance with PRISMA guidelines and has been registered in PROSPERO (CRD42022300857). A computerized literature search was performed using EMBASE, MEDLINE and CINAHL as well as the Cochrane Central register of trials from database inception to May 2022. Randomized controlled trials (RCTs) were included. A random effects network meta-analysis within a frequentist setting was performed for the primary outcome live birth rate/ongoing pregnancy rate (LBR)/(OPR) and for multiple pregnancy rate. Participants/materials, setting, methods Couples/single women undergoing one or more cycles of stimulated IUI for various indications using partner’s or donor sperm were included. The effect sizes of the outcome were estimated as odds ratio (ORs) and presented along with their 95% confidence interval (CIs). We used network plots to illustrate head-to-head comparisons. The superiority of the interventions was assessed; we calculated the probability of being the best, the mean rank, and the surface under cumulative ranking (SUCRA). Main results and the role of chance 57 RCTs were identified comparing stimulation protocols. These included oral agents (clomiphene, letrozole, anastrozole, tamoxifene, estrogen), injectable hormones (FSH, hMG +/- GnRH agonists and GnRH antagonists), alone or in combination, compared to other stimulation agent (s). The included trials randomized 11341 participants across 14 countries between 1994 and 2020. Fourteen trials were multi-centre and 43 (75%) were single centre. Comparisons were assessed in relation to LBR/OPR and in relation to multiple pregnancy rate in order to identify which combination is superior in achieving maximum success rates without compromising safety. The comparison between ovarian stimulation with gonadotrophins and clomiphene had the most trials and participants and showed significantly higher OPR/LBR for ovarian stimulation with gonadotrophins compared to clomiphene (OR: 1.58, CI 1.24-2.02). No stimulation or stimulation with clomiphene or aromatase inhibitors demonstrated significantly lower OPR/LBR compared to clomiphene + estrogen (OR no stimulation: 0.29, CI 0.10-0.83; OR clomiphene: 0.27, CI 0.10-0.69; OR letrozole 0.30, CI 0.11-0.83) and gonadotrophins + letrozole (OR no stimulation: 0.22, CI 0.06-0.82; OR clomiphene: 0.21, CI 0.06-0.73; OR letrozole 0.23, CI 0.06-0.85).The safety of each ovarian stimulation strategy was assessed in terms of risk of multiple pregnancy. Limitations, reasons for caution Comparing different stimulation protocols for IUI is challenging in view of heterogeneity amongst included trials. Different add-ons could benefit specific patient groups but based on the available data, safe recommendations cannot be proposed according to subfertility diagnosis. Various add-ons, protocol variations and sperm preparation techniques could also affect outcomes. Wider implications of the findings Stimulation strategies aiming to increase success rates should be safe in terms of multiple pregnancy. Strict cancellation criteria and monitoring are required. Cost effectiveness, ease of administration, chances of cycle cancellation and couple’s preferences should be taken into consideration. Future trials should take into account subfertility background and patient characteristics. Trial registration number not applicable

A Bayesian-adaptive decision-theoretic approach can reduce the sample sizes for multiarm exercise oncology trials

ObjectivesAdaptive designs may reduce trial sample sizes and costs. This study illustrates a Bayesian-adaptive decision-theoretic design applied to a multiarm exercise oncology trial. Study Design and SettingIn the Physical exercise during Adjuvant Chemotherapy Effectiveness Study (PACES) trial, 230 breast cancer patients receiving chemotherapy were randomized to supervised resistance and aerobic exercise (OnTrack), home-based physical activity (OncoMove) or usual care (UC). Data were reanalyzed as an adaptive trial using both Bayesian decision-theoretic and a frequentist group-sequential approach incorporating interim analyses after every 36 patients. Endpoint was chemotherapy treatment modifications (any vs. none). Bayesian analyses were performed for different continuation thresholds and settings with and without arm dropping and both in a ‘pick-the-winner’ and a ‘pick-all-treatments-superior-to-control’ setting. ResultsTreatment modifications occurred in 34% of patients in UC and OncoMove vs. 12% in OnTrack (P = 0.002). Using a Bayesian-adaptive decision-theoretic design, OnTrack was identified as most effective after 72 patients in the ‘pick-the-winner’ setting and after 72–180 patients in the ‘pick-all-treatments-superior-to-control’ setting. In a frequentist setting, the trial would have been stopped after 180 patients, and the proportion of patients with treatment modifications was significantly lower for OnTrack than UC. ConclusionA Bayesian-adaptive decision-theoretic approach substantially reduced the sample size required for this three-arm exercise trial, especially in the ‘pick-the-winner’ setting.

Paricalcitol and Extended Release Calcifediol for Treatment of Secondary Hyperparathyroidism in Non-Dialysis Chronic Kidney Disease: Results From a Network Meta-Analysis.

Secondary hyperparathyroidism (SHPT) is a complication of chronic kidney disease (CKD) affecting mineral and bone metabolism and characterized by excessive parathyroid hormone (PTH) production and parathyroid hyperplasia. The objective of this analysis was to compare the efficacy and adverse effects of extended release calcifediol (ERC) and paricalcitol (PCT) by assessing their effect on biomarkers PTH, calcium and Phosphate in patients with non-dialysis CKD (ND CKD). A systematic literature research (SLR) was performed in PubMed to identify randomized control trials (RCTs). Quality assessment was done with the GRADE method. The effects of ERC vs PCT were compared using random effects in a frequentist setting. Nine RCTs comprising 1426 patients were included in the analyses. The analyses were performed on two overlapping networks, due to non-reporting of outcomes in some of the included studies. No head-to-head trials were identified. No statistically significant differences in PTH reduction were found between PCT and ERC. Treatment with PCT showed statistically significant increases in calcium compared to ERC (0.2 mg/dl increase, 95% CI: -0.37 to -0.05 mg/dl). No differences in effects on phosphate were observed. This NMA showed that ERC is comparable in lowering PTH levels vs PCT. ERC displayed avoidance of potentially clinically relevant increases in serum calcium, offering an effective and well tolerated treatment option for the management of SHPT in patients with ND CKD.

Feel-Good Thompson Sampling for Contextual Bandits and Reinforcement Learning

Thompson Sampling has been widely used for contextual bandit problems due to the flexibility of its modeling power. However, a general theory for this class of methods in the frequentist setting is still lacking. In this paper, we present a theoretical analysis of Thompson Sampling, with a focus on frequentist regret bounds. In this setting, we show that the standard Thompson Sampling is not aggressive enough in exploring new actions, leading to suboptimality in some pessimistic situations. A simple modification called Feel-Good Thompson Sampling, which favors high reward models more aggressively than the standard Thompson Sampling, is proposed to remedy this problem. We show that the theoretical framework can be used to derive Bayesian regret bounds for standard Thompson Sampling, and frequentist regret bounds for Feel-Good Thompson Sampling. It is shown that in both cases, we can reduce the bandit regret problem to online least squares regression estimation. For the frequentist analysis, the online least squares regression bound can be directly obtained using online aggregation techniques which have been well studied. The resulting bandit regret bound matches the minimax lower bound in the finite action case. Moreover, the analysis can be generalized to handle a class of linearly embeddable contextual bandit problems (which generalizes the popular linear contextual bandit model). The obtained result again matches the minimax lower bound. Finally we illustrate that the analysis can be extended to handle some MDP problems.

Designing truncated priors for direct and inverse Bayesian problems

The Bayesian approach to inverse problems with functional unknowns, has received significant attention in recent years. An important component of the developing theory is the study of the asymptotic performance of the posterior distribution in the frequentist setting. The present paper contributes to the area of Bayesian inverse problems by formulating a posterior contraction theory for linear inverse problems, with truncated Gaussian series priors, and under general smoothness assumptions. Emphasis is on the intrinsic role of the truncation point both for the direct as well as for the inverse problem, which are related through the modulus of continuity as this was recently highlighted by Knapik and Salomond (2018).

Evaluation of operational system structures based on reliability data

AbstractSystems of components have a structure that plays an important role in determining how the reliability of the individual components relates to the reliability of the system. The system reliability can be computed from component reliabilities using results from basic probability theory in the simplest case with all of the components assumed to act independently of one another. However, in the case of dependence, such calculations can be much more involved. When reliability data have been independently collected on both the system and each component in the system, it can be difficult to model any possible dependence between components. Established methods use the known structure of a system, along with these data, to assess whether the reliability of the individual components are mutually independent. In this paper, we expand this methodology to include an assessment of the type of dependence that may exist between the components. This is based on finding the system structure that would most likely produce the observed reliability data, under independence. In the frequentist setting, the likelihood approach is used to find these structures and an observed confidence measure is used to assess the strength of the statistical evidence in favor of each possible structure. In the Bayesian setting, posterior probabilities along with Bayes factors are used. An example demonstrates how these methods can be used in an applied setting.

Sampling properties of the Bayesian posterior mean with an application to WALS estimation

Many statistical and econometric learning methods rely on Bayesian ideas. When applied in a frequentist setting, their precision is often assessed using the posterior variance. This is permissible asymptotically, but not necessarily in finite samples. We explore this issue focusing on weighted-average least squares (WALS), a Bayesian-frequentist ‘fusion’. Exploiting the sampling properties of the posterior mean in the normal location model, we derive estimators of the finite-sample bias and variance of WALS. We study the performance of the proposed estimators in an empirical application and a closely related Monte Carlo experiment which analyze the impact of legalized abortion on crime.

A Bayesian dose-response meta-analysis model: A simulations study and application.

Dose–response models express the effect of different dose or exposure levels on a specific outcome. In meta-analysis, where aggregated-level data is available, dose–response evidence is synthesized using either one-stage or two-stage models in a frequentist setting. We propose a hierarchical dose–response model implemented in a Bayesian framework. We develop our model assuming normal or binomial likelihood and accounting for exposures grouped in clusters. To allow maximum flexibility, the dose–response association is modelled using restricted cubic splines. We implement these models in R using JAGS and we compare our approach to the one-stage dose–response meta-analysis model in a simulation study. We found that the Bayesian dose–response model with binomial likelihood has lower bias than the Bayesian model with normal likelihood and the frequentist one-stage model when studies have small sample size. When the true underlying shape is log–log or half-sigmoid, the performance of all models depends on choosing an appropriate location for the knots. In all other examined situations, all models perform very well and give practically identical results. We also re-analyze the data from 60 randomized controlled trials (15,984 participants) examining the efficacy (response) of various doses of serotonin-specific reuptake inhibitor (SSRI) antidepressant drugs. All models suggest that the dose–response curve increases between zero dose and 30–40 mg of fluoxetine-equivalent dose, and thereafter shows small decline. We draw the same conclusion when we take into account the fact that five different antidepressants have been studied in the included trials. We show that implementation of the hierarchical model in Bayesian framework has similar performance to, but overcomes some of the limitations of the frequentist approach and offers maximum flexibility to accommodate features of the data.

Anesthetic efficacy in vital asymptomatic teeth using different local anesthetics: a systematic review with network meta-analysis.

ObjectivesThis study aimed to evaluate the efficacy of various local anesthesia (LA) in vital asymptomatic teeth.Materials and MethodsRandomized controlled trials comparing pulpal anesthesia of various LA on vital asymptomatic teeth were included in this review. Searches were conducted in the Cochrane CENTRAL, MEDLINE (via PubMed), EMBASE, ClinicalTrials.gov, Google Scholar and 3 field-specific journals from inception to May 3, 2019. Study selection, data extraction, and risk of bias assessment using Cochrane Risk of Bias Tool were done by 2 independent reviewers in duplicate. Network meta-analysis (NMA) was performed within the frequentist setting using STATA 15.0. The LA was ranked, and the surface under the cumulative ranking (SUCRA) line was plotted. The confidence of the NMA estimates was assessed using the CINeMA web application.ResultsThe literature search yielded 1,678 potentially eligible reports, but only 42 were included in this review. For maxillary buccal infiltration, articaine 4% with epinephrine 1:100,000 was more efficacious than lidocaine 2% with epinephrine 1:100,000 (odds ratio, 2.11; 95% confidence interval, 1.14–3.89). For mandibular buccal infiltration, articaine 4% with epinephrine 1:100,000 was more efficacious than various lidocaine solutions. The SUCRA ranking was highest for articaine 4% with epinephrine when used as maxillary and mandibular buccal infiltrations, and lidocaine 2% with epinephrine 1:80,000 when used as inferior alveolar nerve block. Inconsistency and imprecision were detected in some of the NMA estimates.ConclusionsArticaine 4% with epinephrine is superior when maxillary or mandibular infiltration is required in vital asymptomatic teeth.

Immunogenicity and Safety of Ebola Virus Vaccines in Healthy Adults’ Volunteers: A Systematic Review and Network Meta-Analysis

Background: Clinical development of Ebola virus vaccines was accelerated by the West African Ebola virus epidemic which remain the deadliest in history ever witnessed. We aimed to compare and rank Ebola virus vaccines according to their immunogenicity and safety. Methods: We conducted a network meta-analysis using random-effect model within a frequentist setting to obtain direct and indirect comparison of trials. We searched Medline, Embase, and the Cochrane library (CENTRAL) until March 31, 2019, for published phase I or II randomized controlled trials (RCTs) vaccine against Ebola virus in healthy adult’s volunteers. We used the Cochrane Risk of Bias Tools to evaluate RCTs methods, and we extracted the proportion or number of participants who seroconverted and reported adverse events. The primary outcome was immunogenicity assessed by the seroconversion rates, defined as number of participants in each group who showed at least a four-fold increase in antibody titer for enzyme-linked immunosorbent assay (ELISA) 28 days after vaccination. Findings: A total of 18 RCTs, which assessed 18 different treatments involving 58 comparisons and 3,633 participants were included in the analysis. The mean age of participants was between 18 and 65 years, and 59% of them were male. The overall quality of evidence was rated as moderate to low. For immunogenicity, 2,198 (61%) participants had seroconversion at 28-days post-vaccination, and the recombinant vesicular virus-based vaccine expressing a Zaire Ebolavirus glycoprotein (rVSVΔ-ZEBOV-GP) at dose of 2 × 107 was significantly the better vaccine (P-score 0·78). Among the 4,746 reported adverse event (AEs), 97 were severe. For the injection-site pain, a total of 1,222 AEs was reported, and Placebo was associated with few events (P-score 0·93). For fatigue (805 reported AEs) and for headache (1,103 reported AEs), the DNA-EBOV vaccine at dose of ≤ 4 mg was the best one with a P-score of 0·96 and 0·85 respectively. For myalgia, 759 AEs were reported, and the ChAd3 vaccine at dose of 1010 was associated with a lower risk (P-score 0·95). Among the 760 AEs reported for fever, the rVSV-ZEBOV vaccine at dose of ≤ 5 × 105 was ranked as the best one (P-score 0·78). Except for the fatigue AEs, no significant global heterogeneity was found. Interpretation: To prevent the next outbreak of Ebola virus disease, the rVSVΔ-ZEBOV-GP vaccine at dose of 2 × 107 appears as the best available option for sufficiently high immunogenicity. Ad5.ZEBOV (1·6 × 1011) might be a possible alternative for immunogenicity. For AEs, no evidence of the best one vaccine is emerged· Larger RCTs are need to clarify the best candidate vaccine in terms of immunogenicity and safety. Funding Statement: None. Declaration of Interests: The authors declare no competing interests in relation with this work. Ethics Approval Statement: The study protocol number registered in the International Register of Systematic Reviews (PROSPERO) was: CRD42018109473.

Sampling Properties of the Bayesian Posterior Mean With an Application to WALS Estimation

Many statistical and econometric learning methods rely on Bayesian ideas, often applied or reinterpreted in a frequentist setting. Two leading examples are shrinkage estimators and model averaging estimators, such as weighted-average least squares (WALS). In many instances, the accuracy of these learning methods in repeated samples is assessed using the variance of the posterior distribution of the parameters of interest given the data. This may be permissible when the sample size is large because, under the conditions of the Bernstein--von Mises theorem, the posterior variance agrees asymptotically with the frequentist variance. In finite samples, however, things are less clear. In this paper we explore this issue by first considering the frequentist properties (bias and variance) of the posterior mean in the important case of the normal location model, which consists of a single observation on a univariate Gaussian distribution with unknown mean and known variance. Based on these results, we derive new estimators of the frequentist bias and variance of the WALS estimator in finite samples. We then study the finite-sample performance of the proposed estimators by a Monte Carlo experiment with design derived from a real data application about the effect of abortion on crime rates.

Effectiveness of treatments for acute and sub-acute mechanical non-specific low back pain: protocol for a systematic review and network meta-analysis

BackgroundNon-specific low back pain (LBP) is the leading cause of disability worldwide. Acute LBP usually has a good prognosis, with rapid improvement within the first 6 weeks. However, the majority of patients develop chronic LBP and suffer from recurrences. For clinical management, a plethora of treatments is currently available but evidence of the most effective options is lacking. The objective of this study will be to identify the most effective interventions to relieve pain and reduce disability in acute and sub-acute non-specific LBP.Methods/designWe will search electronic databases (MEDLINE, Embase, CENTRAL) from inception onwards. The eligible population will be individuals with non-specific LBP older than 18 years, both males and females, who experience pain less than 6 weeks (acute) or between 6 and 12 weeks (subacute). Eligible interventions and comparators will include all conservative rehabilitation or pharmacological treatments provided by any health professional; the only eligible study design will be a randomized controlled trial. The primary outcomes will be pain intensity and back-specific functional status. Secondary outcomes will be any adverse events. Studies published in languages other than English will also potentially be included. Two reviewers will independently screen the titles and abstracts retrieved from a literature search, as well as potentially relevant full-text articles. General characteristics, potential effect modifiers, and outcome data will be extracted from the included studies, and the risk of bias will be appraised. Conflicts at all levels of screening and abstraction will be resolved through team discussions. After describing the results of the review, if appropriate, a random effects meta-analysis and network meta-analysis will be conducted in a frequentist setting, assuming equal heterogeneity across all treatment comparisons and accounting for correlations induced by multi-arm studies using a multivariate normal model.DiscussionOur systematic review will address the uncertainties in the use of pharmacological or non-pharmacological treatments, and their relative efficacy, for acute and subacute LBP. These findings will be useful for patients, healthcare providers, and policymakers.Systematic review registrationPROSPERO CRD42018102527

Studentized bootstrap model-averaged tail area intervals.

In many scientific studies, the underlying data-generating process is unknown and multiple statistical models are considered to describe it. For example, in a factorial experiment we might consider models involving just main effects, as well as those that include interactions. Model-averaging is a commonly-used statistical technique to allow for model uncertainty in parameter estimation. In the frequentist setting, the model-averaged estimate of a parameter is a weighted mean of the estimates from the individual models, with the weights typically being based on an information criterion, cross-validation, or bootstrapping. One approach to building a model-averaged confidence interval is to use a Wald interval, based on the model-averaged estimate and its standard error. This has been the default method in many application areas, particularly those in the life sciences. The MA-Wald interval, however, assumes that the studentized model-averaged estimate has a normal distribution, which can be far from true in practice due to the random, data-driven model weights. Recently, the model-averaged tail area Wald interval (MATA-Wald) has been proposed as an alternative to the MA-Wald interval, which only assumes that the studentized estimate from each model has a N(0, 1) or t-distribution, when that model is true. This alternative to the MA-Wald interval has been shown to have better coverage in simulation studies. However, when we have a response variable that is skewed, even these relaxed assumptions may not be valid, and use of these intervals might therefore result in poor coverage. We propose a new interval (MATA-SBoot) which uses a parametric bootstrap approach to estimate the distribution of the studentized estimate for each model, when that model is true. This method only requires that the studentized estimate from each model is approximately pivotal, an assumption that will often be true in practice, even for skewed data. We illustrate use of this new interval in the analysis of a three-factor marine global change experiment in which the response variable is assumed to have a lognormal distribution. We also perform a simulation study, based on the example, to compare the lower and upper error rates of this interval with those for existing methods. The results suggest that the MATA-SBoot interval can provide better error rates than existing intervals when we have skewed data, particularly for the upper error rate when the sample size is small.

Mixtures of multivariate restricted skew-normal factor analyzer models in a Bayesian framework

The mixture of factor analyzers (MFA) model, by reducing the number of free parameters through its factor-analytic representation of the component covariance matrices, is an important statistical model to identify hidden or latent groups in high dimensional data. Recent approaches to extend the approach to skewed data or skewness in the latent groups have been examined in a frequentist setting where there are some known computational limitations. For these reasons we consider a Bayesian approach to the restricted skew-normal mixtures of factor analysis MFA model. We examine the performance and flexibility of the approach on real datasets and illustrate some of the computational advantages in a missing data setting.

Comparative efficacy of treatments for Clostridium difficile infection: a systematic review and network meta-analysis

Several new treatments for Clostridium difficile infections have been investigated. We aimed to compare and rank treatments for non-multiply recurrent infections with C difficile in adults. We did a random effects network meta-analysis within a frequentist setting to obtain direct and indirect comparisons of trials. We searched MEDLINE, Embase, Web of Science, Cochrane Central Register of Controlled Trials, and ClinicalTrials.gov for published and unpublished trials from the creation of these databases until June 30, 2017. We included randomised controlled trials of treatments for non-multiply recurrent infections with confirmed C difficile in adults (at least 18 years) that reported both primary cure and recurrence rates, and we used the Cochrane Risk of Bias tool to appraise trial methods. For our analysis, we extracted the total numbers of patients with primary cure and recurrence from published and unpublished reports. The primary outcome was sustained symptomatic cure, defined as the number of patients with resolution of diarrhoea minus the number with recurrence or death. Of 23 004 studies screened, 24 trials, which comprised 5361 patients and 13 different treatments, were included in the analysis. The overall quality of evidence was rated as moderate to low. For sustained symptomatic cure, fidaxomicin (odds ratio 0·67, 95% CI 0·55-0·82) and teicoplanin (0·37, 0·14-0·94) were significantly better than vancomycin. Teicoplanin (0·27, 0·10-0·70), ridinilazole (0·41, 0·19-0·88), fidaxomicin (0·49, 0·35-0·68), surotomycin (0·66, 0·45-0·97), and vancomycin (0·73, 0·56-0·95) were better than metronidazole. Bacitracin was inferior to teicoplanin (0·22, 0·06-0·77) and fidaxomicin (0·40, 0·17-0·94), and tolevamer was inferior to all drugs except for LFF571 (0·50, 0·18-1·39) and bacitracin (0·67, 0·28-1·58). Global heterogeneity of the entire network was low (Cochran's Q=15·70; p=0·47). Among the treatments for non-multiply recurrent infections by C difficile, the highest quality evidence indicates that fidaxomicin provides a sustained symptomatic cure most frequently. Fidaxomicin is a better treatment option than vancomycin for all patients except those with severe infections with C difficile and could be considered as a first-line therapy. Metronidazole should not be recommended for treatment of C difficile. None.

Penalized indirect inference

Parameter estimates of structural economic models are often difficult to interpret at the light of the underlying economic theory. Bayesian methods have become increasingly popular as a tool for conducting inference on structural models since priors offer a way to exert control over the estimation results. Similarly to Bayesian estimation, this paper proposes a penalized indirect inference estimator that allows researchers to obtain economically meaningful parameter estimates in a frequentist setting. The asymptotic properties of the estimator are established for both correctly and incorrectly specified models, as well as under strong and weak parameter identification. A Monte Carlo study reveals the role of the penalty function in shaping the finite sample distribution of the estimator. The advantages of using this estimator are highlighted in the empirical study of a state-of-the-art dynamic stochastic general equilibrium model.

BRISC: bootstrap for rapid inference on spatial covariances

In geostatistics, inference on spatial covariance parameters of the Gaussian process is often critical to scientists for understanding structural dependence in data. Finite‐sample inference customarily proceeds either using posterior distributions from fully a Bayesian approach or via resampling/subsampling techniques in a frequentist setting. Resampling methods, in particular, the bootstrap, have become more attractive in the modern age of big data as, unlike Bayesian models that require sequential sampling from Markov chain Monte Carlo, they naturally lend themselves to parallel computing resources. However, a spatial bootstrap involves an expensive Cholesky decomposition to decorrelate the data. In this manuscript, we develop a highly scalable parametric spatial bootstrap that uses sparse Cholesky factors for parameter estimation and decorrelation. The proposed bootstrap for rapid inference on spatial covariances (BRISC) algorithm requires linear memory and computations and is embarrassingly parallel, thereby delivering substantial scalability. Simulation studies highlight the accuracy and computational efficiency of our approach. Analysing large satellite temperature data, BRISC produces inference that closely matches that delivered from a state‐of‐the‐art Bayesian approach, while being several times faster. The R package BRISC is now available for download from GitHub (https://github.com/ArkajyotiSaha/BRISC) and will be available on CRAN soon. Copyright © 2018 John Wiley & Sons, Ltd.