Correct Inference Research Articles

Optimum data collection and fusion schemes in WBSN

In this paper first we define our strategies aiming at minimising number of communicated data while keeping data integrity in wireless body sensor networks (WBSNs). In this way, we introduce modified Fisher test, develop Spline interpolation as the behaviour function and define controlling parameters. To achieve at significant results, we propose three efficient algorithms. Furthermore, at the second step, we represent our method to calculate the priority of vital sign data packets. For this purpose, we employ Spline interpolation function and define six new controlling parameters. At the third step, for correct inference of patient's situations we introduce our method which develops adaptive neuro fuzzy inference system (ANFIS) with cross-validation. Simulation results in MATLAB R2018b demonstrate the optimum performance of our schemes for number of communicated data, network lifetime, priority based data communication and also correct inference of patient's situations.

Inference Under Random Limit Bootstrap Measures

Asymptotic bootstrap validity is usually understood as consistency of the distribution of a bootstrap statistic, conditional on the data, for the unconditional limit distribution of a statistic of interest. From this perspective, randomness of the limit bootstrap measure is regarded as a failure of the bootstrap. We show that such limiting randomness does not necessarily invalidate bootstrap inference if validity is understood as control over the frequency of correct inferences in large samples. We first establish sufficient conditions for asymptotic bootstrap validity in cases where the unconditional limit distribution of a statistic can be obtained by averaging a (random) limiting bootstrap distribution. Further, we provide results ensuring the asymptotic validity of the bootstrap as a tool for conditional inference, the leading case being that where a bootstrap distribution estimates consistently a conditional (and thus, random) limit distribution of a statistic. We apply our framework to several inference problems in econometrics, including linear models with possibly nonstationary regressors, CUSUM statistics, conditional Kolmogorov–Smirnov specification tests and tests for constancy of parameters in dynamic econometric models.

Autistic traits influence the strategic diversity of information sampling: Insights from two-stage decision models.

Information sampling can reduce uncertainty in future decisions but is often costly. To maximize reward, people need to balance sampling cost and information gain. Here we aimed to understand how autistic traits influence the optimality of information sampling and to identify the particularly affected cognitive processes. Healthy human adults with different levels of autistic traits performed a probabilistic inference task, where they could sequentially sample information to increase their likelihood of correct inference and may choose to stop at any moment. We manipulated the cost and evidence associated with each sample and compared participants’ performance to strategies that maximize expected gain. We found that participants were overall close to optimal but also showed autistic-trait-related differences. Participants with higher autistic traits had a higher efficiency of winning rewards when the sampling cost was zero but a lower efficiency when the cost was high and the evidence was more ambiguous. Computational modeling of participants’ sampling choices and decision times revealed a two-stage decision process, with the second stage being an optional second thought. Participants may consider cost in the first stage and evidence in the second stage, or in the reverse order. The probability of choosing to stop sampling at a specific stage increases with increasing cost or increasing evidence. Surprisingly, autistic traits did not influence the decision in either stage. However, participants with higher autistic traits inclined to consider cost first, while those with lower autistic traits considered cost or evidence first in a more balanced way. This would lead to the observed autistic-trait-related advantages or disadvantages in sampling optimality, depending on whether the optimal sampling strategy is determined only by cost or jointly by cost and evidence.

Bibliographic measures of top-tier finance and information systems journals

Purpose Pay, tenure and promotion decisions are frequently based upon inferences regarding the value of faculty research. Meanwhile, departmental, college and university reputations are frequently based on perceptions regarding the quality of research being produced by its faculty. Making correct inferences requires accurate measurement of research quality, which is often based upon the journal through which results are shared. This research expands upon the research found elsewhere through its detailed investigation of leading journals in two business disciplines, including examination of four different citation-based measures and four journal characteristics which are exogenous to the quality of any individual piece of research. The paper aims to discuss this issue. Design/methodology/approach This study assists in the development of an accurate perspective regarding research quality, by studying the popular Journal Citation Reports (JCR) impact factor. A further expansion on the past literature is consideration of three newer journal quality metrics: SCImago Journal Rank (SJR), Source Normalized Impact per Paper (SNIP) and percentage of articles cited. Top-tier journals in finance and information systems are compared to evaluate the consistency of these measures across disciplines. Differences in journal characteristics and their impact on citation-rate based measures of quality are also examined. The potential impact of discipline-based variation in acceptance rate, issue frequency, the time since journal inception and total reviewers are put forth as additional potential exogenous factors that may impact the perception of journal quality. t-Tests are employed for discipline comparisons, while correlation and multiple regression are used for journal characteristic analysis. Findings There is a significant difference in the JCR impact measures of high-quality finance journals vs high-quality information systems journals, which are correlated with a variety of journal-specific factors including the journal’s acceptance rate and frequency of issue. Information systems journals domination of finance journals persists whether one considers mean, median, minimum or maximum impact factors. SJR measures for finance journals are consistently higher than information systems journals, though the SJR value of any individual journal can be quite volatile. By comparison, the SNIP metric rates premier information systems journals higher. Over 12 percent more of the articles in leading information systems journals are cited during the initial three years. Research limitations/implications Logical extensions of this research include examining journals in other business disciplines. One could also evaluate quality measures reaction to variation in journal characteristics (i.e. changes in acceptance rates). Furthermore, one could include other measures of journal quality, including the recently released CiteScore metric. Such research will build on the present research and improve the accuracy of research quality assessment. Practical implications To the extent that citation-based research measures and journal-specific factors vary across disciplines as demonstrated by our investigation, discipline-specific traits should be considered adjusted for, when making inferences about the long-term value of recently published research. For instance, finance faculty publishing in journals with JCR readings of 2.0 are in journals that are 53 percent above the discipline’s average, while information systems faculty publishing in journals with JCR readings of 2.0 are in journals that are 18 percent below the discipline’s average. Furthermore, discipline-specific differences in journal characteristics, leading to differences in citation-based quality measures, should be considered when making inferences about the long-term value of recently published research in the process of making recommendations regarding salary adjustments, retention and promotion. Social implications Quantity and quality of research are two hallmarks of leading research institutions. Assessing research quality is very problematic because its definition has changed from being based on review process (i.e. blind refereed), to acceptance rates, to impact factors. Furthermore, the impact factor construct has been a lightning rod of controversy as researchers, administrators and journals themselves argue over which metric to employ. This research is attempting to assess how impact factors and journal characteristics may influence the impact factors, and how these interactions vary business discipline. The research is especially important and relevant to the authors which separately chair departments including finance and information systems faculty, and therefore are in roles requiring assessment of faculty research productivity including quality. Originality/value This study is a detailed analysis of bibliographic aspects of the top-tier journals in two quantitative business areas. In addition to the popular JCR, SJR and SNIP measures of performance, the analysis studies the seldom-examined percentage of the article cited metric. A deeper understanding of citation-based measures is obtained though the evaluation of changes in how journals have been rated on these metrics over time. The research shows that there are discipline-related systematic differences in both citation-based research measures and journal-specific factors and that these discipline-specific traits should be considered when making inferences about the long-term value of recently published research. Furthermore, discipline-specific difference in journal characteristics, leading to differences in citation-based quality measures, should be considered when making personnel and remuneration decisions.

Estimating permanent price impact via machine learning

In this paper, we show that vector auto-regression (VAR) models, which are commonly used to estimate permanent price impact, are misspecified and can produce conflicting and incorrect inferences when the price impact function is nonlinear. We propose an alternative method to estimate permanent price impact by modifying a reinforcement learning (RL) framework. Our approach assumes the data is stationary and Markov, but is otherwise unrestrictive. We obtain empirical estimates for our model using an iterative learning rule and demonstrate that our model captures nonlinearities and makes correct inferences.

Estimating Permanent Price Impact via Machine Learning

In this paper, we show that vector auto-regression (VAR) models, which are commonly used to estimate permanent price impact, are misspecified and can produce conflicting and incorrect inferences when the price impact function is nonlinear. We propose an alternative method to estimate permanent price impact by modifying a reinforcement learning (RL) framework. Our approach assumes the data is stationary and Markov, but is otherwise unrestrictive. We obtain empirical estimates for our model using an iterative learning rule and demonstrate that our model captures nonlinearities and makes correct inferences.

LARGE MULTIPLE GRAPHICAL MODEL INFERENCE VIA BOOTSTRAP.

Large economic and financial networks may experience stage-wise changes as a result of external shocks. To detect and infer a structural change, we consider an inference problem within a framework of multiple Gaussian Graphical Models when the number of graphs and the dimension of graphs increase with the sample size. In this setting, two major challenges emerge as a result of the bias and uncertainty inherent in the regularization required to treat such overparameterized models. To deal with these challenges, the bootstrap method is utilized to approximate the sampling distribution of a likelihood ratio test statistic. We show theoretically that the proposed method leads to a correct asymptotic inference in a high-dimensional setting, regardless of the distribution of the test statistic. Simulations show that the proposed method compares favorably to its competitors such as the Likelihood Ratio Test. Finally, our statistical analysis of a network of 200 stocks reveals that the interacting units in the financial network become more connected as a result of the financial crisis between 2007 and 2009. More importantly, certain units respond more strongly than others. Furthermore, after the crisis, some changes weaken, while others strengthen.

Inference of temperature and density profiles via forward modeling of an x-ray imaging crystal spectrometer within the Minerva Bayesian analysis framework.

At the Wendelstein 7-X stellarator, the X-ray imaging crystal spectrometer provides line integrated measurements of ion and electron temperatures, plasma flows, as well as impurity densities from a spectroscopic analysis of tracer impurity radiation. In order to infer the actual profiles from line integrated data, a forward modeling approach has been developed within the Minerva Bayesian analysis framework. In this framework, the inversion is realized on the basis of a complete forward model of the diagnostic, including error propagation and utilizing Gaussian processes for generation and inference of arbitrary shaped plasma parameter profiles. For modeling of line integrated data as measured by the detector, the installation geometry of the spectrometer, imaging properties of the crystal, and Gaussian detection noise are considered. The inversion of line integrated data is achieved using the maximum posterior method for plasma parameter profile inference and a Markov chain Monte Carlo sampling of the posterior distribution for calculating uncertainties of the inference process. The inversion method shows a correct and reliable inference of temperature and impurity density profiles from synthesized data within the estimated uncertainties along the whole plasma radius. The application to measured data yields a good match of derived electron temperature profiles to data of the Thomson scattering diagnostic for central electron temperatures between 2 and 5 keV using argon impurities.

Approximating faces of marginal polytopes in discrete hierarchical models

The existence of the maximum likelihood estimate in a hierarchical log-linear model is crucial to the reliability of inference for this model. Determining whether the estimate exists is equivalent to finding whether the sufficient statistics vector $t$ belongs to the boundary of the marginal polytope of the model. The dimension of the smallest face $\mathbf{F}_{t}$ containing $t$ determines the dimension of the reduced model which should be considered for correct inference. For higher-dimensional problems, it is not possible to compute $\mathbf{F}_{t}$ exactly. Massam and Wang (2015) found an outer approximation to $\mathbf{F}_{t}$ using a collection of submodels of the original model. This paper refines the methodology to find an outer approximation and devises a new methodology to find an inner approximation. The inner approximation is given not in terms of a face of the marginal polytope, but in terms of a subset of the vertices of $\mathbf{F}_{t}$. Knowing $\mathbf{F}_{t}$ exactly indicates which cell probabilities have maximum likelihood estimates equal to $0$. When $\mathbf{F}_{t}$ cannot be obtained exactly, we can use, first, the outer approximation $\mathbf{F}_{2}$ to reduce the dimension of the problem and then the inner approximation $\mathbf{F}_{1}$ to obtain correct estimates of cell probabilities corresponding to elements of $\mathbf{F}_{1}$ and improve the estimates of the remaining probabilities corresponding to elements in $\mathbf{F}_{2}\setminus\mathbf{F}_{1}$. Using both real-world and simulated data, we illustrate our results, and show that our methodology scales to high dimensions.

Accurate inferences of others' thoughts depend on where they stand on the empathic trait continuum

This research explores the possibility that a person's (perceiver's) prospects of making a correct inference of another person's (target's) inner states depends on the personal characteristics of the target, potentially relating to how readable they are. Twenty-seven targets completed the Empathy Quotient (EQ) and were classified as having low, average or high EQ. They were unobtrusively videoed while thinking of an event of happiness, gratitude, anger and sadness. After observing targets thinking of such a past event, fifty-two perceivers (participants) in Study 1 were asked to infer what the target was thinking, and fifty perceivers in Study 2 were asked to rate the target's expression – positive or negative. Results suggested that (1) perceivers' accuracy in detecting targets' thoughts depended on which EQ group the target belonged to, and (2) target readability is not a proxy measure for level of target expressiveness. In other words, something about EQ status renders targets more or less easy to read in a way that is not simply explained by expressive people being more readable. We conclude with discussion of the importance of the target's trait as well as situation they experience in determining how accurately a perceiver might infer their inner states.

Solving linear regression without skewness of the residuals’ distribution

Linear ordinary least squares (OLS) regression assumes an unskewed distribution of the residuals for correct inference and prediction. A proof is given that for Manly’s exponential transformation of the dependent variable, there is always at least one solution for such that the skewness of the standardized residuals’ distribution is zero. A computer code in Mathematica, together with an illustrative example, are provided. Generalized linear models are discussed briefly in comparison.

How can we improve inferences from surveys? A new look at the convertible debt questions from the Graham and Harvey survey data

We revisit the survey questions on convertible bond issue motives from the influential study of Graham and Harvey (2001). Our question-conditional analysis connecting survey answers with firm characteristics reveals that the conclusions on two of the four convertible debt theories from the original study need to be revised. More particularly, the delayed equity rationale and the sequential financing rationale on convertible bond issuance do not receive support from the question-conditional tests. Our results indicate that adding a question-conditional analysis is essential in obtaining correct inferences from survey data.

Eggshell resorption, and embryonic mobilization and accumulation of calcium and metals in eggs of wild and captive Capercaillies Tetrao urogallus

Although it is fairly well known that eggshells are the major source of Ca and other micronutrients for developing avian embryos, potential anatomical and/or ecological correlates explaining the variation in embryo-induced depletion of micronutrients from eggshells remain to be addressed. Using unhatched eggs at various developmental stages and post-hatched eggshells of wild and captive-bred birds of a large precocial species, the Capercaillie Tetrao urogallus, we explored two major questions. (i) Do egg size, regional eggshell thickness as well as levels of two major micronutrients (Ca and Mg) and eight trace elements (Cr, Cu, Mn, Fe, Co, Cd, Pb and Zn) measured in egg shells and contents vary with embryonic age within and between wild and captive-bred bird eggs? (ii) How the proportions of eggshell elements become depleted during the entire process of embryogenesis? The eggshells of wild Capercaillies were thinner at the equator and the sharp pole, and contained more Mg and Ca, while in their contents there was more Cu, Cd and Pb compared with captive birds. All ten metal concentrations measured in the egg contents increased (from 15% to 285% more) with the age of embryos. The concentrations of six elements measured in eggshells (Cu, Mn, Fe, Co, Cd and Zn) decreased consistently (from 15 to 70% less) during embryogenesis. This suggests the presence of these elements in the innermost decalcified shell layer and their further mobilization into the embryos/egg contents. Our results strongly imply that most trace elements present in the egg contents/developing embryos are mobilized from the eggshells. We highlight the fact that knowledge of the magnitude of embryogenesis-related functional changes in the physical (such as shell thickness) and chemical traits of the eggs and eggshells of oviparous vertebrates is fundamental for drawing correct inferences regarding the response of an individual to variable environmental conditions.

Molecular analysis of the RHD pseudogene by duplex real-time polymerase chain reaction.

In this study, we aimed to present a strategy for the detection of the RHD pseudogene (RHDψ) based on a real-time polymerase chain reaction (PCR) assay. The D-negative phenotype is associated with many genetic alterations. In populations with African ancestry, this phenotype commonly results from the silent variant RHDψ. The evaluation of RHDψ is essential for correct inference of the RhD phenotype in order to avoid false-positive results. We utilised a new method for the simultaneous detection of RHDψ and a fragment from exon 5 of the wild-type RHD gene based on duplex real-time PCR assay. The PCR assay allowed specific detection of RHDψ. There was complete agreement between the results generated by the new test and the results generated by molecular analysis performed using end-point PCR methods previously described. The assay developed is easy to execute and presents the potential for routine use at blood banks and other associated facilities where it is desired to determine the presence of RHDψ.

A weighting approach to making inference with probabilistically linked data

Record linkage is the act of bringing together records from two files that are believed to belong to the same unit (e.g., a person or business). It is a low‐cost way of increasing the set of variables available for analysis. Errors may arise in the linking process if an error‐free unit identifier is not available. Two types of linking errors include an incorrect link (records belonging to two different units are linked) and a missed record (an unlinked record for which a correct link exists). Naively ignoring linkage errors may mean that analysis of the linked file is biased. This paper outlines a “weighting approach” to making correct inference about regression coefficients and population totals in the presence of such linkage errors. This approach is designed for analysts who do not have the expertise or time to use specialist software required by other approaches but who are comfortable using weights in inference. The performance of the estimator is demonstrated in a simulation study.

Satellite-detected forest disturbance forecasts American marten population decline: The case for supportive space-based monitoring

Limited monitoring resources often constrain rigorous monitoring practices to species or populations of conservation concern. Insufficient monitoring can induce a tautology as lack of monitoring resources makes it difficult to determine whether a species or population deserves additional monitoring resources. When in-situ monitoring resources are limited, remote habitat monitoring could be a useful supplementary tool, as linking parameterized species distribution models to spatially explicit time-series of environmental correlates allows iterative prediction of population change. Yet the performance of predictive forecasts or hindcasts has been difficult to evaluate. We paired contemporary field data, historic population estimates, and a remotely-sensed archive of landscape change to evaluate predictions of American marten (Martes americana) population decline owing to habitat loss in Maine, USA. We estimated contemporary spatial patterns in marten density relative to landscape disturbance with spatial capture-recapture models. We compared current density estimates to historical density calculations to evaluate population decline, and compared historical calculations to habitat-based model predictions to evaluate the efficacy of habitat monitoring as a proxy for direct monitoring. Marten density was negatively associated with the proportion of surrounding regenerating forest, and point estimates within focal townships were 50–80% lower than historical calculations. Habitat-based hindcasts of marten density across our entire focal area interest suggested a smaller population decline (roughly 50%) within our focal area. Thus, although habitat-based predictions underpredicted marten decline, they provided correct directional inference. Habitat monitoring and predictions from species distribution models may provide useful inference about population changes given trends in habitat at limited expense when in-situ information is lacking.

Estimation of linear dynamic panel data models with time‐invariant regressors

SummaryWe present a sequential approach to estimating a dynamic Hausman–Taylor model. We first estimate the coefficients of the time‐varying regressors and subsequently regress the first‐stage residuals on the time‐invariant regressors. In comparison to estimating all coefficients simultaneously, this two‐stage procedure is more robust against model misspecification, allows for a flexible choice of the first‐stage estimator, and enables simple testing of the overidentifying restrictions. For correct inference, we derive analytical standard error adjustments. We evaluate the finite‐sample properties with Monte Carlo simulations and apply the approach to a dynamic gravity equation for US outward foreign direct investment.

A transition model for analyzing multivariate longitudinal data using Gaussian copula approach

Longitudinal studies often involve multiple mixed response variables measured repeatedly over time. Although separate modeling of these multiple mixed response variables can be easily performed, they may lead to inefficient estimates and consequently, misleading inferences. For obtaining correct inference, one needs to model multiple mixed responses jointly. In this paper, we use copula models for defining a multivariate distribution for multiple mixed outcomes at each time point. Then, we use transition model for considering association between longitudinal measurements. Two simulation studies are performed for illustration of the proposed approach. The results of the simulation studies show that the use of the separate models instead of the joint modeling leads to inefficient parameter estimates. The proposed approach is also used for analyzing two real data sets. The first data set is a part of the British Household Panel Survey. In this data set, annual income and life satisfaction are considered as the continuous and the ordinal correlated longitudinal responses, respectively. The second data set is a longitudinal data about heart failure patients. This study is a treatment–control study, where the effect of a treatment is simultaneously investigated on readmission and referral to doctor as two binary associated longitudinal responses.

Inference in Moment Inequality Models That Is Robust to Spurious Precision under Model Misspecification

Standard tests and confidence sets in the moment inequality literature are not robust to model misspecification in the sense that they exhibit spurious precision when the identified set is empty. This paper introduces tests and confidence sets that provide correct asymptotic inference for a pseudo-true parameter in such scenarios, and hence, do not suffer from spurious precision.

Inference in Moment Inequality Models That Is Robust to Spurious Precision under Model Misspecification

Standard tests and confidence sets in the moment inequality literature are not robust to model misspecification in the sense that they exhibit spurious precision when the identified set is empty. This paper introduces tests and confidence sets that provide correct asymptotic inference for a pseudo-true parameter in such scenarios, and hence, do not suffer from spurious precision.