Inconsistent Estimates Research Articles

Probabilistic deterioration modeling of bridge component condition with random effects

ABSTRACT Timely maintenance of bridge components is critical for bridge management functions. With reliable deterioration models, highway agencies can efficiently allocate funding for bridge maintenance and customize maintenance schedules to meet agency budgets. The increased public expectation of acceptable levels of service for bridges coupled with other competing needs makes it crucially important to accurately estimate bridge future conditions so that adequate resources may be allocated for repair and reconstruction purposes. Accurately predicting bridge condition is challenging due to the inherent random nature of factors impacting deterioration, the existence of unobserved variables that are not measured, panel nature of the data and the effects of bridge-specific correlation. Without accounting for these factors, the resulting estimated deterioration models may have biased and inconsistent parameter estimates. This article assembled a comprehensive set of bridge and climate data from the National Bridge Inventory (NBI) and the South Carolina Climatology office. Bridge component deterioration models for bridges on state highways in South Carolina were estimated using an ordered probit model with random effects specification to account for the randomness and panel nature of the bridge data. The study results are useful for various bridge management tasks including maintenance programming, budgeting and bridge asset evaluation.

Artificial Intelligence Design for Ship Structures: A Variant Multiple-Input Neural Network-Based Ship Resistance Prediction

Abstract In this work, we have developed a data-driven artificial intelligence (AI) solution to assist the ship hull design process. Specifically, we have developed and implemented an AI-based multiple-input neural network model to realize the real-time prediction of the total resistance of the ship hull structure while avoiding the inconsistent estimates from different types of design input parameters. It is demonstrated that the developed AI-based machine learning algorithm as a prediction tool can assist the ship hull design process by accurately providing the total resistance of ship hulls in real time. Moreover, we have conducted design tasks to validate the proposed method, and the validation results show that a well-trained artificial neural network model can avoid the problem of different sensitivities due to the different degrees of influence of the input parameters on the output parameter. The proposed AI-based data-driven solution provides a real-time hydrodynamic performance calculation, which can predict the hyperdynamic performances of ship hulls based on their geometry modification parameters. This approach gives a consistent prediction in terms of accuracy when facing different geometry modification parameters, and it in turn provides a fast and accurate AI-based method to assist ship hull design to achieve an optimum forecast accuracy in the entire design space, making an advance to artificial intelligence assist design in naval architecture engineering.

A Cell-to-Pack State Estimation Extension Method Based on a Multilayer Difference Model for Series-Connected Battery Packs

Battery degradation and cell inconsistency challenge the existing state estimation methods, which often ignore the influence of cell inconsistency or suffer from heavy computation complexity. To address the difficulty of state estimation by cell inconsistency and realize joint estimation of the state-of-charge (SOC) and capacity for series-connected battery packs, a novel cell-to-pack state estimation extension method based on a multilayer difference model (MDM) is investigated. The proposed extension method can efficiently realize accurate SOC and capacity estimation for a battery pack based on the existing estimation algorithms for a single cell. Two state-difference estimators for SOC and capacity differences are constructed and realized separately through adaptive extended Kalman filter and recursive least squares algorithm. Considering that time-varying characteristics of states and state differences are different, the MDM estimator runs in multiple timescales. Based on battery pack cycling experiments, the cell-to-cell consistency evolution during aging is revealed. The proposed MDM’s accuracy, efficiency, and adaptability are verified through the experiments of a series-connected battery pack under different dynamic conditions. The proposed method also shows adaptability to various battery temperatures and different cell inconsistencies. The maximum SOC root-mean-squared error and capacity average error for battery pack during aging are 2.35% and 3.02%, respectively.

Security and Privacy Protection of Internet of Vehicles Consensus Algorithm Based on Wireless Sensors

Due to its large network scale, open communication environment, unstable wireless network, and other characteristics, it is extremely vulnerable to attacks and causes security problems, resulting in the collapse of the Internet of Vehicles system. The application of the Internet of Vehicles is becoming more and more extensive, but there are still problems such as information security and privacy leakage in the Internet of Vehicles. Through the analysis of the security threats and privacy protection requirements faced by the Internet of Vehicles system, this paper mainly studies information security, vehicle identity privacy, and location privacy in the process of Internet of Vehicles wireless communication. Therefore, it is urgent to conduct research on the information security and privacy protection issues of the Internet of Vehicles. This paper discusses the research on the security and privacy protection of the consensus algorithm for the Internet of Vehicles based on wireless sensors, compares and analyzes the wireless sensor data privacy protection protocols based on sharding technology, Tongtai encryption technology, and perturbation technology, and selects an optimized Kalman consensus filter. The algorithm is applied to the node information exchange of the sensor network, and two filters (low pass and band pass) are used to unify the observations and covariance of the network. Estimation of the sensor network state with and without data packet loss, the effect of system estimation error under different packet loss rates, data privacy protection algorithm performance, vehicle network data communication volume, and confusion factors on algorithm efficiency and the node energy consumption was compared and analyzed. Based on the application of wireless sensors, the estimation error and inconsistency estimation error of the algorithm in this paper finally converge to about 0.5, and both can maintain good stability and have good robustness. In addition, the communication volume of the algorithm in this paper is about 30% of the SCPDA algorithm. The Kalman consensus filtering algorithm reduces the amount of confusing data sent, improves privacy protection, and also achieves lower communication overhead.

Estimation of Dynamic Panel Data Regression Parameters Using Generalized Methods of Moment

Panel data is a combination of cross section and time series. There are two panel data models, namely static and dynamic panel data. Because seeing the advantages of the dynamic panel data model which is able to overcome endogeneity problems related to the use of the dependent variable lag where in the static panel data model the use of the dependent variable lag causes the estimation results to be biased and inconsistent, so the author examines the dynamic panel data regression model. In the dynamic data model there is a lag of the dependent variable, this variable is correlated with error. Thus, estimation using OLS will result in a biased and inconsistent estimator. To overcome this, the dynamic panel data model can be estimated using the GMM Blundell-Bond approach. Based on the discussion, the parameter estimation formula for dynamic panel data regression using the Blundell-Bond GMM approach is as follows:

Infinite diameter confidence sets in Hedges’ publication bias model

Meta-analysis, the statistical analysis of results from separate studies, is a fundamental building block of science. But the assumptions of classical meta-analysis models are not satisfied whenever publication bias is present, which causes inconsistent parameter estimates. Hedges’ selection function model takes publication bias into account, but estimating and inferring with this model is tough for some datasets. Using a generalized Gleser–Hwang theorem, we show there is no confidence set of guaranteed finite diameter for the parameters of Hedges’ selection model. This result provides a partial explanation for why inference with Hedges’ selection model is fraught with difficulties.

A Guide to Autoregressive Distributed Lag Models for Impulse Response Estimations*

We provide a guide to using autoregressive distributed lag models for impulse response estimations with an identified structural shock or an external instrument for the shock. We illustrate how specifications widely used in practice can lead to inconsistent and inefficient estimators. We further review empirical results from previous papers and show that some results appear to be statistical artefacts. We propose a simple method to avoid such a false conclusion from biases or large standard errors and obtain consistent and precise estimates.

UNDERSTANDING THE UNSETTLED EVIDENCE OF THE EFFECTIVENESS OF SELECTIVE EDUCATION IN THE VALUE-ADDED APPROACH

ABSTRACT This study compares the estimated grammar school effect in different regression models, and explains why previous evidence of the effetiveness of grammar school is mixed. Like most studies of school effectiveness evaluation, previous research on grammar school effect usually applies regression to control for confounding between-school factors and determines whether attending grammar schools is associated with an academic benefit. While this value-added approach is very feasible and widely adopted, there is usually substantial variation in the evidence produced when statistical choices differ. Applying the national data with more than 149,000 pupils, the study presents the sensitivity of grammar school effectiveness when baseline variables, outcome variables, and regression techniques differ. The findings will help researchers understand the mixed evidence of the effectiveness of grammar schools so far, and the unsettled debate on grammar schools nationally. While inconsistent estimations due to passive design is a common trait in school effectiveness evaluation studies, grammar schools’ effectiveness might be even more sensitive to statistical choices due to their special student intakes. Interpreting the variation between different statistical models, the findings caution researchers in dealing with the evidence in grammar school effectiveness and call for stronger designs as well as more carefully selected explanatory variables.

Optimal minimax random designs for weighted least squares estimators

Summary This work studies an experimental design problem where the values of a predictor variable, denoted by $x$, are to be determined with the goal of estimating a function $m(x)$, which is observed with noise. A linear model is fitted to $m(x)$, but it is not assumed that the model is correctly specified. It follows that the quantity of interest is the best linear approximation of $m(x)$, which is denoted by $\ell(x)$. It is shown that in this framework the ordinary least squares estimator typically leads to an inconsistent estimation of $\ell(x)$, and rather weighted least squares should be considered. An asymptotic minimax criterion is formulated for this estimator, and a design that minimizes the criterion is constructed. An important feature of this problem is that the $x$ values should be random, rather than fixed. Otherwise, the minimax risk is infinite. It is shown that the optimal random minimax design is different from its deterministic counterpart, which was studied previously, and a simulation study indicates that it generally performs better when $m(x)$ is a quadratic or a cubic function. Another finding is that, when the variance of the noise goes to infinity, the random and deterministic minimax designs coincide. The results are illustrated for polynomial regression models and a generalization is given in the Supplementary Material.

Functional data analysis for longitudinal data with informative observation times.

In functional data analysis for longitudinal data, the observation process is typically assumed to be noninformative, which is often violated in real applications. Thus, methods that fail to account for the dependence between observation times and longitudinal outcomes may result in biased estimation. For longitudinal data with informative observation times, we find that under a general class of shared random effect models, a commonly used functional data method may lead to inconsistent model estimation while another functional data method results in consistent and even rate-optimal estimation. Indeed, we show that the mean function can be estimated appropriately via penalized splines and that the covariance function can be estimated appropriately via penalized tensor-product splines, both with specific choices of parameters. For the proposed method, theoretical results are provided, and simulation studies and a real data analysis are conducted to demonstrate itsperformance.

StarBeast3: Adaptive Parallelized Bayesian Inference under the Multispecies Coalescent.

As genomic sequence data become increasingly available, inferring the phylogeny of the species as that of concatenated genomic data can be enticing. However, this approach makes for a biased estimator of branch lengths and substitution rates and an inconsistent estimator of tree topology. Bayesian multispecies coalescent (MSC) methods address these issues. This is achieved by constraining a set of gene trees within a species tree and jointly inferring both under a Bayesian framework. However, this approach comes at the cost of increased computational demand. Here, we introduce StarBeast3—a software package for efficient Bayesian inference under the MSC model via Markov chain Monte Carlo. We gain efficiency by introducing cutting-edge proposal kernels and adaptive operators, and StarBeast3 is particularly efficient when a relaxed clock model is applied. Furthermore, gene-tree inference is parallelized, allowing the software to scale with the size of the problem. We validated our software and benchmarked its performance using three real and two synthetic data sets. Our results indicate that StarBeast3 is up to one-and-a-half orders of magnitude faster than StarBeast2, and therefore more than two orders faster than *BEAST, depending on the data set and on the parameter, and can achieve convergence on large data sets with hundreds of genes. StarBeast3 is open-source and is easy to set up with a friendly graphical user interface. [Adaptive; Bayesian inference; BEAST 2; effective population sizes; high performance; multispecies coalescent; parallelization; phylogenetics.]

Improving Penalized Logistic Regression Model with Missing Values in High-Dimensional Data

Analysis without adequate handling of missing values may lead to inconsistent and biased estimates. Despite multiple imputations becoming a widely used approach in handling missing data, manuscript researchers generally encounter missing data in their respective studies. In high-dimensional data, penalized regression is a popular technique for performing feature selection and coefficient estimation simultaneously. However, one of the most vital issues with high-dimensional data is that it often contains large quantities of missing data that common multiple imputation approaches may not work correctly. Therefore, this study uses imputations penalized regression models as an extension of the penalized methods to improve the performance and impute missing values in high-dimensional data. The method was applied to real-life high-dimensional datasets for the different number of features, sample sizes, and missing dataset rates to evaluate its efficiency. The method was also compared with other existing imputation penalized methods for high-dimensional data. The comparative experimental results indicate that the proposed method outperforms its competitors by achieving higher sensitivity, specificity, and classification accuracy values.

Comparing Single- and Multiple-Question Designs of Measuring Family Income in China Family Panel Studies

Family income questions in general purpose surveys are usually collected with either a single-question summary design or a multiple-question disaggregation design. It is unclear how estimates from the two approaches agree with each other. The current paper takes advantage of a large-scale survey that has collected family income with both methods. With data from 14,222 urban and rural families in the 2018 wave of the nationally representative China Family Panel Studies, we compare the two estimates, and further evaluate factors that might contribute to the discrepancy. We find that the two estimates are loosely matched in only a third of all families, and most of the matched families have a simple income structure. Although the mean of the multiple-question estimate is larger than that of the single-question estimate, the pattern is not monotonic. At lower percentiles up till the median, the single-question estimate is larger, whereas the multiple-question estimate is larger at higher percentiles. Larger family sizes and more income sources contribute to higher likelihood of inconsistent estimates from the two designs. Families with wage income as the main income source have the highest likelihood of giving consistent estimates compared with all other families. In contrast, families with agricultural income or property income as the main source tend to have very high probability of larger single-question estimates. Omission of certain income components and rounding can explain over half of the inconsistencies with higher multiple-question estimates and a quarter of the inconsistencies with higher single-question estimates.

The Role of Conditional Likelihoods in Latent Variable Modeling

In psychometrics, the canonical use of conditional likelihoods is for the Rasch model in measurement. Whilst not disputing the utility of conditional likelihoods in measurement, we examine a broader class of problems in psychometrics that can be addressed via conditional likelihoods. Specifically, we consider cluster-level endogeneity where the standard assumption that observed explanatory variables are independent from latent variables is violated. Here, “cluster” refers to the entity characterized by latent variables or random effects, such as individuals in measurement models or schools in multilevel models and “unit” refers to the elementary entity such as an item in measurement. Cluster-level endogeneity problems can arise in a number of settings, including unobserved confounding of causal effects, measurement error, retrospective sampling, informative cluster sizes, missing data, and heteroskedasticity. Severely inconsistent estimation can result if these challenges are ignored.

New estimation for heteroscedastic single-index measurement error models

In this paper, we propose a novel estimation for heteroscedastic single-index models with covariate measurement errors, in which both the conditional mean and conditional variance functions of the response given the covariates have a single-index structure. When covariates are directly observable, we show that the index parameter vector can be estimated consistently by the estimation obtained by fitting a misspecified linear quantile regression model under some mild regularity conditions. It is well known that naively treating mismeasured covariates as error-free usually leads to inconsistent estimators. To account for measurement errors in covariates, we establish a new estimation procedure based on corrected quantile loss function, and obtain the asymptotic consistency and normality of the resulting estimators. Finally, the finite sample performance of the proposed estimation method is illustrated by simulation studies and an empirical analysis of a real dataset.

Neural Network Training Using Closed-Loop Data: Hazards and an Instrumental Variable (IVNN) Solution

An increasing trend in the use of neural networks in control systems is being observed. The aim of this paper is to reveal that the straightforward application of learning neural network feedforward controllers with closed-loop data may introduce parameter inconsistency that degrades control performance, and to provide a solution. The proposed method employs instrumental variables to ensure consistent parameter estimates. A nonlinear system example reveals that the developed instrumental variable neural network (IVNN) approach asymptotically recovers the optimal solution, while pre-existing approaches are shown to lead to inconsistent estimates.

A Novel Channel Inconsistency Estimation Method for Azimuth Multichannel SAR Based on Maximum Normalized Image Sharpness

For azimuth multi-channel synthetic aperture radar (SAR), unavoidable inconsistency errors between channels can degrade SAR image quality severely, leading to possible ghost targets and image defocusing, etc. To address this issue, a novel channel inconsistency estimation method is proposed based on maximum normalized image sharpness. First, channel amplitude and time delay errors are corrected in the coarse compensation step. Then images of each channel are attained by azimuth spectrum recovery and imaging processing. Next, range-variant channel phase errors are estimated via optimizing normalized image sharpness, which reaches the maximum value when the image is focused well or ghost targets are suppressed completely. The Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm is employed to get the optimal solution based on the derived gradient of objective function. Finally, the ultimate image is formed through adding up phase compensated images of each channel. By optimizing the focused image quality, the proposed algorithm achieves high estimation accuracy. Simulated data and real multi-channel SAR data are processed to demonstrate the effectiveness of the proposed method.

Multiple Improvements of Multiple Imputation Likelihood Ratio Tests

Multiple imputation (MI) inference handles missing data by imputing the missing values $m$ times, and then combining the results from the $m$ complete-data analyses. However, the existing method for combining likelihood ratio tests (LRTs) has multiple defects: (i) the combined test statistic can be negative, but its null distribution is approximated by an $F$-distribution; (ii) it is not invariant to re-parametrization; (iii) it fails to ensure monotonic power owing to its use of an inconsistent estimator of the fraction of missing information (FMI) under the alternative hypothesis; and (iv) it requires nontrivial access to the LRT statistic as a function of parameters instead of data sets. We show, using both theoretical derivations and empirical investigations, that essentially all of these problems can be straightforwardly addressed if we are willing to perform an additional LRT by stacking the $m$ completed data sets as one big completed data set. This enables users to implement the MI LRT without modifying the complete-data procedure. A particularly intriguing finding is that the FMI can be estimated consistently by an LRT statistic for testing whether the $m$ completed data sets can be regarded effectively as samples coming from a common model. Practical guidelines are provided based on an extensive comparison of existing MI tests. Issues related to nuisance parameters are also discussed.

Scene Adaptive Phase Inconsistency Estimation for Multi-channel ScanSAR System

Operating multi-channel synthetic aperture radar (SAR) in ScanSAR mode enables an ultrawide-swath and high-resolution imagery. A critical step in multi-channel SAR data processing is to accurately estimate and compensate phase inconsistency between channels. However, the Doppler spectrum distribution of ScanSAR relies heavily on the characteristic of illuminated regions, which may make the traditional time-domain phase estimation method failure under nonuniform scenes. In order to ensure that the performance of phase estimation is robust with respect to the variation of scenes, an innovative scene adaptive phase inconsistency estimation method for the ScanSAR system is presented in this letter, which can estimate channel phase error more accurately by extracting scene-center frequency from received data. The performance of the presented approach is evaluated using real airborne multi-channel SAR data. Experimental results show that the modified phase inconsistency estimation method is more effective and adaptable to different scenes compared with the conventional method.

Self-Avoiding Trail model in two-dimensional regular lattices: Study of conformational quantities and relation with the Self-Avoiding Walk model

The Self-Avoiding Trail (SAT) is a random walk model represented by an N-step lattice path, with forbidden step overlaps. The SAT is a variant of the Self-Avoiding Walk (SAW) model, where a site is visited only once. There is a consensus that SAT is in the same universality class of SAW, implying that the scaling behavior of conformational quantities, represented by the exponents ν0 and Δ1, is the same in both models. However, in comparison to SAW, SAT results exhibit most pronounced finite-size effects and inconsistent estimates of ν0, depending on the lattice coordination number and conformational quantity. The scaling behavior of the persistence length, λN, is in disagreement with recent estimates for the SAW model. Also, the influence of coordination number on the lattice-dependent constants requires a better understanding. In this study, we revisit the SAT model with data generated by the pivot algorithm up to N=8000 steps, for the hexagonal, square and triangular lattices. Accordingly, we find compelling evidences that ν0 and Δ1 values are the same for the SAT and SAW models, regardless the lattice and conformational quantity used, while the most pronounced finite-size effect comes from the higher amplitude of the correction to scaling terms. Similarly to SAW, we find the λN convergence to lattice dependent constant λ∞(SAW) for the SAT model. Based on the ratio λ∞(SAW)/λ∞(SAT)∝z/2, where z is the lattice coordination number, we find that the factor z/2 quantifies the relation between SAW and SAT lattice dependent constants.