This paper shows that the critical values for likelihood ratio inference of the threshold point in Hansen (2000) are too large if we restrict the confidence set as an interval, which can partially explain why Hansen’s confidence interval is conservative when the threshold effect is not too small. We provide appropriate critical values and show that different from conventional critical values, these new critical values are invariant to the structural change in the error and covariate distributions.

Prolonged diabetic retinopathy (DR), glaucoma, and age-related macular degeneration (AMD) may lead to vision loss. Hence, early detection and treatment are crucial to prevent irreversible vision loss. Fundus retinal images have been widely used to help detect these diseases. Manual screening is susceptible to human errors, tedious, and expensive. Hence, artificial intelligence (AI) techniques have been widely employed to overcome these constraints. This paper reviewed the work published on automated retinal health detection models using various machine learning (ML) and deep learning (DL) techniques. We reviewed 142 papers and 262 studies (124 on glaucoma, 60 on AMD, and 78 on DR) from January 2012 to June 2024 using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We found that Convolutional Neural Networks (CNN) and Support Vector Machines (SVM) models were widely used in DL and ML techniques, respectively. To the best of our knowledge, this is the first review developed for detecting AMD, DR, and glaucoma using AI techniques over the last decade. We have discussed the limitations of the present methods and also suggested future directions for accurately detecting eye diseases.

This paper studies control function (CF) approaches in endogenous threshold regression where the threshold variable is allowed to be endogenous. We first use a simple example to show that the structural threshold regression (STR) estimator of the threshold point in Kourtellos, Stengos and Tan (2016, Econometric Theory 32, 827–860) is inconsistent unless the endogeneity level of the threshold variable is low compared to the threshold effect. We correct the CF in the STR estimator to generate our first CF estimator using a method that extends the two-stage least squares procedure in Caner and Hansen (2004, Econometric Theory 20, 813–843). We develop our second CF estimator which can be treated as an extension of the classical CF approach in endogenous linear regression. Both these approaches embody threshold effect information in the conditional variance beyond that in the conditional mean. Given the threshold point estimates, we propose new estimates for the slope parameters. The first is a by-product of the CF approach, and the second type employs generalized method of moment (GMM) procedures based on two new sets of moment conditions. Simulation studies, in conjunction with the limit theory, show that our second CF estimator and confidence interval for the threshold point together with the associated second GMM estimator and confidence interval for the slope parameter dominate the other methods. We further apply the new estimation methodology to an empirical application from international trade to illustrate its usefulness in practice.

Background Hypertension is a significant global disease burden. Mobile health (mHealth) offers a promising means to provide patients with hypertension with easy access to health care services. Yet, its efficacy needs to be validated, especially in lower-income areas with a high-salt diet. Objective This study aims to assess the efficacy of an mHealth app–based intervention in supporting patients’ self-management of hypertension. Methods A 2-arm randomized controlled trial was conducted among 297 patients with hypertension at the General Hospital of Ningxia Medical University, Ningxia Hui Autonomous Region, China. Participants selected via convenience sampling were randomly allocated into intervention and control groups. Intervention group participants were trained and asked to use an mHealth app named Blood Pressure Assistant for 6 months. They could use the app to record and upload vital signs, access educational materials, and receive self-management reminders and feedback from health care providers based on the analysis of the uploaded data. Control group participants received usual care. Blood pressure (BP) and 2 questionnaire surveys about hypertension knowledge and lifestyle behavior were used to assess all participants at baseline and 6 months. Data analysis was performed with SPSS software using 2-tailed t tests and a chi-square test. Results There were no significant differences in baseline characteristics and medication use between the 2 groups (all P>.05). After 6 months, although both groups show a significant pre-post improvement (P<.001 each), the BP control rate (ie, the proportion of patients with a systolic BP of <140 mm Hg and diastolic BP of <90 mm Hg) in the intervention group was better than that in the control group (100/111, 90.1% vs 75/115, 65.2%; P<.001). The mean systolic and diastolic BP were significantly reduced by 25.83 (SD 8.99) and 14.28 (SD 3.74) mm Hg in the intervention group (P<.001) and by 21.83 (SD 6.86) and 8.87 (SD 4.22) mm Hg in the control group (P<.001), respectively. The differences in systolic and diastolic BP between the 2 groups were significant (P<.001 and P=.01, respectively). Hypertension knowledge significantly improved only in the intervention group in both pre-post and intergroup comparisons (both P<.001). However, only intragroup improvement was observed for lifestyle behaviors in the intervention group (P<.001), including medication adherence (P<.001), healthy diet (P=.02), low salt intake (P<.001), and physical exercises (P=.02), and no significant difference was observed in the control group or on intergroup comparisons. Conclusions This research shows that the mHealth app–based intervention has the potential to improve patient health knowledge and support self-management among them toward a healthier lifestyle, including medication adherence, low-salt diets, and physical exercises, thereby achieving optimal BP control. Further research is still needed to verify the specific effects of these interventions. Trial Registration Chinese Clinical Trial Registry ChiCTR1900026437; https://www.chictr.org.cn/showproj.html?proj=38801

Software testing plays a very important role in the software development process. Automated test generation tools increase the effectiveness and efficiency of software testing, and alleviate the problem of low efficiency caused by writing hand-crafted test cases. However, different test case generation methods vary in the size, code coverage, and fault detection capacity of the automatically-produced test suites. Automated test case generation tool based on random testing, Randoop as a representative, randomly and incrementally generates a large number of method sequences, which gives various possible combinations of calling methods, but the size of the test suite is not proportional to test quality. Therefore, there exists a lot of redundancy in the test cases. This paper proposes Randoop-TSR, an approach for identifying and eliminating redundant test cases on the basis of Randoop to improve the process of test generation. Our approach adopts three strategies to realize the removal of redundancy, namely: (i) similarity-based input sequence selection; (ii) redundant and duplicate assert statements elimination based on test smell detection; (iii) redundant test cases elimination without breaking test requirements (i.e., code coverage and mutation score). Randoop-TSR can eliminate redundancy effectively, and greatly reduce the size of test suites and execution time. Furthermore, our approach improves the efficiency and understandability of test cases while retaining code coverage and mutation score.

In social and behavioral sciences, the mediation test based on the indirect effect is an important topic. There are many methods to assess intervening variable effects. In this paper, we focus on the difference method and the product method in mediation models. Firstly, we analyze the regression functions in the simple mediation model, and provide an expectation-consistent condition. We further show that the difference estimator and the product estimator are numerically equivalent based on the least-squares regression regardless of the error distribution. Secondly, we generalize the equivalence result to the three-path model and the multiple mediators model, and prove a general equivalence result in a class of restricted linear mediation models. Thirdly, we investigate the empirical distributions of the indirect effect estimators in the simple mediation model by simulations, and show that the indirect effect estimators are normally distributed as long as one multiplicand of the product estimator is large. Finally, we introduce some popular R packages for mediation analysis and also provide some useful suggestions on how to correctly conduct mediation analysis.

There has been a steady increase in international students pursuing postgraduate coursework education in English speaking countries. Like first-year undergraduate students, these international students need assistance transitioning into the new educational environment and preparing for self-directed, collaborative learning throughout their careers. Drawing on the social constructivist pedagogical approaches, we developed learning tasks that foster self-regulation and collaboration among postgraduate coursework IT students, aligning these tasks with the learning outcomes of the subject Information Design and Content Management. This paper presents the rationale and method for the design of the learning tasks, and how these learning tasks to not only align with the subject learning outcomes but also facilitate self-regulation. A study involving preand post-subject surveys and interviews with 133 subject students will provide us with further insights into the effectiveness of the learning task design and the areas for improvement.

In recent years, intelligent fault diagnosis algorithms using deep learning method have achieved much success. However, the signals collected by sensors contain a lot of noise, which will have a great impact on the accuracy of the diagnostic model. To address this problem, we propose a one-dimensional convolutional neural network with multi-scale kernels (MSK-1DCNN) and apply this method to bearing fault diagnosis. We use a multi-scale convolution structure to extract different fault features in the original signal, and use the ELU activation function instead of the ReLU function in the multi-scale convolution structure to improve the anti-noise ability of MSK-1DCNN; then we use the training set with pepper noise to train the network to suppress overfitting. We use the Western Reserve University bearing data to verify the effectiveness of the algorithm and compare it with other fault diagnosis algorithms. Experimental results show that the improvements we proposed have effectively improved the diagnosis performers of MSK-1DCNN under strong noise and the diagnosis accuracy is higher than other comparison algorithms.

Summary Aeromagnetic exploration is an efficient and convenient geophysical exploration method. Aeromagnetic interference compensation is the key link of aeromagnetic data processing. Current methods for aeromagnetic compensation are mostly multiple linear regression compensation. But the regression ability of variables with strong correlation or non-linear data is poor. A neural network nonlinear compensation model that the training parameters were not determined by standard FOM flight was once proposed. Therefore, the original neural network compensation training model was no longer applicable. To address this issue, we propose a 9-parameter compensation method based on a multilayer BP neural network. In order to make the structure of the neural network suitable for aeromagnetic compensation, the optimal structure is determined through analyzing the node transfer function, training function, number of hidden layers and the number of nodes in each hidden layer. By analyzing the aeromagnetic interference model for optimizing the training parameters and improving the compensation accuracy, 9 compensation training parameters are determined. The experimental results show that the method is effective.

Aiming at the needs and existing problems of the course construction of Modern Communication Theory, this paper introduces the necessity and feasibility of the application of modern educational technology, and make teaching practices by applying modern educational technology in the course of Modern Communication Theory, such as multimedia, classroom management interactive platform, micro-course and MOOC, virtual and simulation software, case, website development, it illustrates the important role of modern educational technology applied in the construction of quality course of Modern Communication Theory and postgraduate training.

We propose three new methods of inference for the threshold point in endogenous threshold regression and two specii¬ cation tests designed to assess the presence of endogeneity and threshold ei¬€ects without necessarily relying on instrumentation of the covariates. The i¬ rst inferential method is a parametric two-stage least squares method and is suitable when instruments are available. The second and third methods are based on smoothing the objective function of the integrated dii¬€erence kernel estimator in dii¬€erent ways and these methods do not require instrumentation. All three methods are applicable irrespective of endogeneity of the threshold variable. The two specii¬ cation tests are constructed using a score-type principle. The threshold ei¬€ect test extends conventional parametric structural change tests to the nonparametric case. A wild bootstrap procedure is suggested to deliver i¬ nite sample critical values for both tests. Simulations show good i¬ nite sample performance of these procedures and the methods provide flexibility in testing and inference for practitioners working with threshold models.

In nonparametric regression, the derivative estimation has attracted much attention in recent years due to its wide applications. In this paper, we propose a new method for the derivative estimation using the locally weighted least absolute deviation regression. Different from the local polynomial regression, the proposed method does not require a finite variance for the error term and so is robust to the presence of heavy-tailed errors. Meanwhile, it does not require a zero median or a positive density at zero for the error term in comparison with the local median regression. We further show that the proposed estimator with random difference is asymptotically equivalent to the (infinitely) composite quantile regression estimator. The proposed method is also extended to estimate the derivatives at the boundaries and to estimate the higher-order derivatives. For the equidistant design, we derive theoretical results for the proposed estimators, including the asymptotic bias and variance, consistency, and asymptotic normality. Finally, we conduct simulation studies to demonstrate that the proposed method has better performance than the existing methods in the presence of outliers and heavy-tailed errors.

In this article, we mainly discuss hypothesis tests to assess whether nonparametric function satisfies monotonicity or convexity in semiparametric partially linear model. For this purpose, we propose a new test statistic for monotonicity or convexity of nonparametric function based on Bernstein polynomials. Furthermore, we employ likelihood ratio statistic to test the significance of regression parameters of the model. We discuss the asymptotic properties of the tests for both nonparametric function and regression parameters. A simulation study is conducted to evaluate the finite sample performance compared with the other methods for shape tests. The method is illustrated by the fuel efficiency study.

The asymptotic distribution of the least squares estimator in threshold regression is expressed in terms of a compound Poisson process when the threshold effect is fixed and as a functional of two-sided Brownian motion when the threshold effect shrinks to zero. This paper explains the relationship between this dual limit theory by showing how the asymptotic forms are linked in terms of joint and sequential limits. In one case, joint asymptotics apply when both the sample size diverges and the threshold effect shrinks to zero, whereas sequential asymptotics operate in the other case in which the sample size diverges first and the threshold effect shrinks subsequently. The two operations lead to the same limit distribution, thereby linking the two different cases. The proofs make use of ideas involving limit theory for sums of a random number of summands.

The existing differenced estimators of error variance in nonparametric regression are interpreted as kernel estimators, and some requirements for a ''good'' estimator of error variance are specified. A new differenced method is then proposed that estimates the errors as the intercepts in a sequence of simple linear regressions and constructs a variance estima-tor based on estimated errors. The new estimator satisfies the requirements for a ''good'' estimator and achieves the asymptotically optimal mean square error. A feasible difference order is also derived, which makes the estimator more applicable. To improve the finite-sample performance, two bias-corrected versions are further proposed. All three es-timators are equivalent to some local polynomial estimators and thus can be interpreted as kernel estimators. To determine which of the three estimators to be used in practice, a rule of thumb is provided by analysis of the mean square error, which solves an open problem in error variance estimation which difference sequence to be used in finite samples. Simulation studies and a real data application corroborate the theoretical results and illustrate the advantages of the new method compared with the existing methods.

This paper shows that when the threshold variable is independent of other covariates, such as in the structural change model, the least squares estimator of the threshold point is consistent even if endogeneity is present.

This paper studies semiparametric efficient estimation of the threshold point in threshold regression. The classical literature of semiparametric efficient estimation rests on the fact that the maximum likelihood estimator is efficient in any parametric submodel for a large class of loss functions. However, in threshold regression, the maximum likelihood estimator is not efficient, while the Bayes estimators are efficient and different loss functions induce different efficient estimators. For an additively separable loss function that separates the efficiency problem of the threshold point from that of other parameters, we show that the semiparametric and parametric efficiency risk bounds coincide. Then we design a semiparametric empirical Bayes estimator to achieve this bound. In consequence, the threshold point can be adaptively estimated even under conditional moment restrictions. We also provide a valid confidence interval called the nonparametric posterior interval for the threshold point. Simulation studies show that the semiparametric empirical Bayes approach is substantially better than existing methods. To illustrate our procedure in practice, we apply it to an economic growth model for detecting different growth patterns.

The regression discontinuity design has become a common framework among applied economists for measuring treatment effects. A key restriction of the existing literature is the assumption that the discontinuity point is known, which does not always hold in practice. This paper extends the applicability of the regression discontinuity design by allowing for an unknown discontinuity point. First, we construct a unified test statistic to check whether there are selection or treatment effects. Our tests are shown to be consistent, and local powers are derived. Also, a bootstrap method is proposed to obtain critical values. Second, we estimate the treatment effect by first estimating the nuisance discontinuity point. It is shown that estimating the discontinuity point does not affect the efficiency of the treatment effect estimator. Simulation studies illustrate the usefulness of our procedures in finite samples.