Probability Interval Research Articles

A Novel Method of Bridge Deflection Prediction Using Probabilistic Deep Learning and Measured Data.

The deflection control of the main girder in suspension bridges, as flexible structures, is critically important during their operation. To predict the vertical deflection of existing suspension bridge girders under the combined effects of stochastic traffic loads and environmental temperature, this paper proposes an integrated deflection interval prediction method based on a Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), a probability density estimation layer, and bridge monitoring data. A time-series training dataset consisting of environmental temperature, vehicle load, and deflection monitoring data was built based on bridge health monitoring data. The CNN-LSTM combined layer is used to capture both local features and long-term dependencies in the time series. A Gaussian distribution (GD) is adopted as the probability density function, and its parameters are estimated using the maximum likelihood method, which outputs the optimal deflection prediction and probability intervals. Furthermore, this paper proposes a method for identifying abnormal deflections of the main girder in existing suspension bridges and establishes warning thresholds. This study indicates that, for short time scales, the CNN-LSTM-GD model achieves a 47.22% improvement in Root Mean Squared Error (RMSE) and a 12.37% increase in the coefficient of determination (R2) compared to the LSTM model. When compared to the CNN-LSTM model, it shows an improvement of 28.30% in RMSE and 6.55% in R2. For long time scales, the CNN-LSTM-GD model shows a 54.40% improvement in RMSE and a 10.22% increase in R2 compared to the LSTM model. Compared to the CNN-LSTM model, it improves RMSE by 38.43% and R2 by 5.31%. This model is instrumental in more accurately identifying abnormal deflections and determining deflection thresholds, making it applicable to bridge deflection early-warning systems.

Simultaneous confidence interval construction for many-to-one of proportion ratios of bilateral correlated data.

In ophthalmology and otolaryngology, data collected from paired body parts are typically reformatted into categorical bilateral data structures for subsequent research. This article applies Donner's equal correlation coefficient model and obtains nine simultaneous confidence intervals (SCI) of proportion ratios under three asymptotic statistical methods and three ways of multiplicity adjustment. The empirical coverage probability and mean interval width are evaluated through Monte Carlo simulations. A real example is used to demonstrate the proposed methods.

Algorithms for Verifying the Correctness of Interval Expert Evaluations

Expert probability is used for decision-making, allows you to assess risks, and is especially useful in conditions of limited and inaccessible objective data. At the same time, there may be contradictions between the estimates, which may complicate analysis and decision-making. This paper describes an algorithm that checks the compliance of expert estimates using probability intervals. One of the sections is devoted to the development of an evaluation method for a new formula based on the original evaluation system. The article describes the user interface that provides interaction with the developed algorithms.

Magnetic signature of deep pit ovens from Paquimé, ceremonial center of the Casas Grandes culture (Northern Mexico)

The Casas Grandes culture in Northern Mexico and Southwest United States has been characterized by its impressive architecture - unique to desert cultures, with houses and constructions of up to four floors built by adobe. In this study, we report an integrated rock-magnetic and archaeomagnetic study from Paquimé archaeological site, considered as an outstanding testimony of the relationship between the cultures of North America and Mesoamerica. The In Situ structures of Paquimé archaeological zone were magnetically analyzed for the first time. Four up to 2.6 m deep Pit Ovens, commonly used for agave cooking, were sampled. 78 specimens were subjected to stepwise alternating field demagnetizations to retrieve characteristic remanence archaeodirections. The great majority of samples yielded evidence of essentially stable, single component remanent magnetization carried by the end members of titanomagnetite solid solutions (almost magnetite phase). Hematite grains apparently also co-exist, but their contribution in total thermoremanence is minor. The archaeomagnetic dating, based on mean declination and inclination, was performed using the global geomagnetic models and the available regional Paleosecular Variation Curves available for Mesoamerica. Additionally, SW United State paleosecular variation patterns from Four Corner region was also considered. The most probable age interval for one of the studied ovens (around 940 AD) is out of Paquimé’s main occupation phase between 1200 and 1450 AD. The age estimation for the PH1 pit oven, around 1150 AD, could be located near the limit of the beginning of the Medio Period. Remained two analyzed ovens yielded up to three possible ages distributed around 1100, 1320 and 1480 AD. Pit Ovens studied here, were long time considered to correspond to same time interval, while these new results attest that these combustion structures encompassed almost six centuries in agreement with the historical fact that the city of Paquimé was abandoned before the arrival of the Spaniards.

A temperature-sensitive points selection method for machine tool based on rough set and multi-objective adaptive hybrid evolutionary algorithm

Reasonable deployment of temperature sensors is the key to accurately monitoring the temperature field of machine tools and improving the accuracy of thermal error prediction and compensation models. To determine the optimal deployment location of sensors, this paper proposes a temperature-sensitive points selection method tightly coupled with rough set and multi-objective optimization. Firstly, the importance of each temperature measurement point to the thermal error is calculated based on the rough set, and information entropy is introduced to amplify the importance difference among adjacent measurement points at the same heat source. Then, with the temperature measurement points groups as the variables, the number of temperature measurement points in the group, and the information importance of the group as the objectives, a multi-objective attribute reduction model is established, which transforms the temperature-sensitive points selection problem into a discrete multi-objective optimization problem. Finally, a multi-objective adaptive hybrid evolutionary algorithm is proposed, which designs a population initialization method based on mutual information and interval probability, and dynamic adaptive evolutionary parameters to achieve optimal temperature-sensitive points selection. Experiments on the high-speed dry hobbing machine verify the superiority and effectiveness of the proposed method.

Short-Term Photovoltaic Power Probabilistic Forecasting Based on Temporal Decomposition and Vine Copula

With the large-scale development of solar power generation, highly uncertain photovoltaic (PV) power output has an increasing impact on distribution networks. PV power generation has complex correlations with various weather factors, while the time series embodies multiple temporal characteristics. To more accurately quantify the uncertainty of PV power generation, this paper proposes a short-term PV power probabilistic forecasting method based on the combination of decomposition prediction and multidimensional variable dependency modeling. First, a seasonal and trend decomposition using a Loess (STL)-based PV time series feature decomposition model is constructed to obtain periodic, trend, and residual components representing different characteristics. For different components, this paper develops a periodic component prediction model based on TimeMixer for multi-scale temporal feature mixing, a long short-term memory (LSTM)-based trend component extraction and prediction model, and a multidimensional PV residual probability density prediction model optimized by Vine Copula optimized with Q-Learning. These components’ results form a short-term PV probabilistic forecasting method that considers both temporal features and multidimensional variable correlations. Experimentation with data from the Desert Knowledge Australia Solar Center (DKASC) demonstrates that the proposed method reduced root mean square error (RMSE) and mean absolute percentage error (MAPE) by at least 14.8% and 22%, respectively, compared to recent benchmark models. In probability interval prediction, while improving accuracy by 4% at a 95% confidence interval, the interval width decreased by 19%. The results show that the proposed approach has stronger adaptability and higher accuracy, which can provide more valuable references for power grid planning and decision support.

Estimation of effective damping ratio for cast-in-place tunnel-form system and evaluation of its role in performance point prediction

The extensive time and computational effort are primary challenges in nonlinear dynamic analysis of tunnel-form concrete systems. These challenges lead engineers to resort to simpler, pushover-based analyses, inherently based on estimating the seismic performance point of the system. Technical literature review indicates that no study has yet rigorously evaluated the accuracy of existing methods for estimating the performance point of tunnel-form systems. To eliminate potential ambiguities, in this study, the seismic performance point of the system under design basis earthquake (with a 475-year return period) has been calculated using three different methods (i.e., displacement coefficient, capacity spectrum, and displacement amplification factor), and compared with the results of accurate nonlinear time-history analysis. In the range of 5-, 7-, and 10-story models studied, the results indicate the inefficiency and insufficiency of the mentioned methods. Investigations reveal that while the capacity spectrum method provides better results, but its process is lengthy, and the displacement coefficient method significantly overestimates the performance point (with more than 80 % error). It was also evident that the displacement amplification factor underestimates the performance point and contradicts the direction of confidence. Based on observations, the use of all three methods for the tunnel-form system requires modifications. The calculated values of effective damping ratio for the tunnel-form system explicitly indicate type A behavior according to the ATC-40 classification. By presenting this parameter in a multi-level format, the shortcomings of both capacity spectrum and displacement coefficient methods are easily addressed. Referring to the results, the calculated value of the displacement amplification factor in the system exceeds the recommended value by the seismic design code, and by adjusting it, satisfactory responses can be obtained in the method based on the displacement amplification factor. Finally, introducing the “probable performance interval” parameter, recommending its use instead of the “performance point” parameter in assessments by pushover analysis is suggested. This parameter is applicable with all three mentioned methods and has been introduced in this study as a desirable factor in compensating for inherent uncertainties related to future earthquakes.

Preoperative cognitive profile predictive of cognitive decline after subthalamic deep brain stimulation in Parkinson's disease.

Cognitive decline represents a severe non-motor symptom of Parkinson's disease (PD) that can significantly reduce the benefits of subthalamic deep brain stimulation (STN DBS). Here, we aimed to describe post-surgery cognitive decline and identify pre-surgery cognitive profile associated with faster decline in STN DBS-treated PD patients. A retrospective observational study of 126 PD patients treated by STN DBS combined with oral dopaminergic therapy followed for 3.54 years on average (SD=2.32) with repeated assessments of cognition was conducted. Pre-surgery cognitive profile was obtained via a comprehensive neuropsychological examination and data analysed using exploratory factor analysis and Bayesian generalized linear mixed models. On the whole, we observed a mild annual cognitive decline of 0.90 points from a total of 144 points in the Mattis Dementia Rating Scale (95% posterior probability interval [-1.19, -0.62]) with high inter-individual variability. However, true score changes did not reach previously reported reliable change cut-offs. Executive deficit was the only pre-surgery cognitive variable to reliably predict the rate of post-surgery cognitive decline. On the other hand, exploratory analysis of electrode localization did not yield any statistically clear results. Overall, our data and models imply mild gradual average annual post-surgery cognitive decline with high inter-individual variability in STN DBS-treated PD patients. Nonetheless, patients with worse long-term cognitive prognosis can be reliably identified via pre-surgery examination of executive functions. To further increase the utility of our results, we demonstrate how our models can help with disentangling true score changes from measurement error in future studies of post-surgery cognitive changes.

An interplay between cross-cultural and psychometric factors in the Montreal Cognitive Assessment: Experience from the language of a small nation

The present study aimed to test the hypothesis that the total word length on the Memory subtest of the Czech version of the MoCA, which is 12 syllables compared to the English version of 7 syllables, would have a significant effect on Delayed Recall scores compared to the newly created well-balanced version of the test (further MoCA-WLE). In the original Czech version of MoCA, we replaced the 12-syllable word list in the Memory subtest with a 7-syllable list (MoCA-WLE) to make it equivalent to the standard English version in this respect. We analyzed data from 83 participants in the original MoCA group (70.63 ± 7.01 years old, 14.61 ± 3.17 years of education, 30.12% males) and 83 participants in the MoCA-WLE group (70.72 ± 6.95 years old, 14.93 ± 3.48 years of education, 30.12% males). We did not find evidence for a significant word-length effect in the original MoCA versus MoCA-WLE Delayed Recall in either the Mann–Whitney U test (W = 3418.0, p = .932) or multilevel binomial regression (b = 0.10, 95% posterior probability interval [−0.46, 0.68]). The present study shows cross-cultural limits in the adaptation of the test material. The results underline the caveats of such an approach to test adaptation. Fortunately, 12-syllables in the MoCA Memory Czech version versus the original 7-syllable list did not show a detectable word-length effect. We did not find evidence for differential item functioning or cultural item bias. The original MoCA Czech version is psychometrically comparable to the original English version.

Clinical manifestations of scarlet fever in children in the megalopolis at the present stage.

BACKGROUND: according to the literature, an increase in severe forms of scarlet fever among children was noted in a number of foreign countries in 2022-2023. AIMS: to identify the current clinical features of scarlet fever in children in a large metropolis of Russia. MATERIALS AND METHODS: an observational descriptive retrospective continuous study of the clinical features of the course of scarlet fever in children was conducted according to 295 case histories of patients diagnosed with Scarlet fever from 2020 to 2022 on the basis of the GBUZ "DGKB named after Z.A. Bashlyaeva DZM", GBUZ "DGKB No. 9 named after G.N. Speransky DZM" in Moscow. The statistical processing of the material uses traditional methods for calculating extensive and intensive indicators (p1 and p2), errors of extensive indicators (m1 and m2) and standard deviations (rmax↔rmip), taking into account the confidence interval (t) and confidence probability (P=95%), a method for comparing indicators using a well-known formula, where at t ≥ 2.0 P ≥ 95% p ≤ 0.05, the definition of standard deviations, taking into account the confidence interval (t) and the confidence probability P = 95%. The comparison of percentages in the analysis of four-field conjugacy tables was performed using the exact Fisher criterion. The differences were considered statistically significant at p 0.05. RESULTS: as a result of the study, it was revealed that the abolition of strict anti-epidemic restrictions on COVID-19 in 2022 led not only to an increase in the incidence of scarlet fever among children from 0 to 18 years of age by 3.6 times in 2022 compared with 2021, but, as a result, to an increase in hospitalization of children in hospitals in Moscow. Thus, the number of hospitalized children with scarlet fever in 2022 increased by 18.0 times compared to 2020. The majority of those hospitalized (83.4%) were children from organized groups, including those aged 3 to 7 years (52.9%) and from 7 to 14 years (39.3%), with an average severity of the disease (93.6%). It was noted that that among the examined children, 53 children (18%) had a history of COVID-19, and among this group of children, one in five children had severe scarlet fever, while among children who did not have a history of COVID-19, these were significantly (p 0.001) 8.4 times less (respectively, 20.7% versus 2.5%). It was revealed that every 8-9 patients (12.2%) had various complications, mainly of a septic nature and those that arose in the first week of the disease, including otitis media, sinusitis, community-acquired pneumonia, parathonsillar and lateral pharyngeal abscesses and toxic shock syndrome. In the second week and later, complications of an infectious and allergic nature were reported, including tubulointerstitial nephritis, acute renal injury, post-infectious cardiomyopathy, acute post-streptococcal glomerulonephritis, toxic allergic vasculitis, post-streptococcal arthritis. Beta-lactam antibiotics remain the etiotropic drugs of choice. Antibiotics of the cephalosporin 53.6% and penicillin 46.4% series were used to treat scarlet fever. CONCLUSIONS: in modern conditions of the megalopolis, scarlet fever occurs with a typical clinical picture. The majority of those hospitalized were children of preschool and school age, only 12.2% of children were diagnosed with various complications of scarlet fever and the vast majority of children (93.6%) had a moderate severity of scarlet fever. It was noted that against the background of an increase in the number of hospitalized patients, the number of patients with severe scarlet fever decreases, while the incidence of complications in patients does not change (12.3% - 12.8%). It was shown that among patients who had suffered from COVID-19, scarlet fever was 8.3 times more likely to occur in severe form compared with children who had no history of COVID-19 disease (respectively, 20.7% vs. 2.5, p 0.0001).

Probability and recurrence interval of large earthquakes (Mw ≥ 6) in the Himalayan seismicity zone

The Himalayan Mountains are one of the world’s most seismically active regions. Due to the active seismic status of the Himalayan belt and ongoing development activities in the region, it is important to keep the seismic hazard estimates updated by incorporating historical and new data. This study uses the Gutenberg-Richter relations for four seismogenic source zones (SSZ) to estimate the probability of earthquakes of a specified magnitude in a given time interval. We used exponential, Weibull, Rayleigh, and log-normal statistical models to assess the probability of earthquakes (Mw ≥ 6). The probable recurrence time of the large earthquake (Mw ≥ 6) can be estimated by maximizing the conditional probability estimates. This recurrence interval is calculated after the elapsed time since the most recent large earthquake in the region. We also apply an autoregressive (AR) forecasting technique to predict earthquake recurrence intervals. The results indicate that the approximate recurrence interval of the large earthquakes (Mw ≥ 6) may be 22.052 ± 5.65 years within the seismicity zones, as per our statistical estimates based on limited and available datasets. The results of this study may be useful for the agencies involved in assessing seismic hazards and developing mitigation and preparedness strategies in the region.

A stochastic constitutive model and its application to HTPB propellant

AbstractTo address the issue of randomness in the mechanical properties of the hydroxyl‐terminated polybutadiene (HTPB) propellant, a stochastic constitutive model (SCM) with a lognormally distributed random parameter Λ was proposed to describe their mechanical behaviors, and the structural integrity of a HTPB propellant grain was analyzed based on it. The results indicate that the stress‐strain curves predicted by the SCM have a good agreement with the experimental curves, and the experimental curves fall within a 95 % probability interval predicted by the SCM. The mechanical response of HTPB propellant grain under ignition pressurization is associated with the random parameters Λ. The maximum equivalent stress and safety factor increase approximately linearly with the increase of random parameters Λ, while the maximum equivalent strain and maximum damage coefficient decrease approximately linearly with the increase of random parameters Λ. The error in the mechanical response of the grain obtained based on the SCM and the experimental constitutive model is basically not more than 2 %, the SCM can effectively characterize the randomness in the mechanical response of propellant grain caused by the dispersion of HTPB propellant mechanical properties.

The impact of green finance on green growth: The role of green energy and green production

This study examines the linkage of green finance and green growth under the regulatory role of green energy and green production in 52 countries worldwide from 2005 to 2019. Applying the Bayesian regression and GMM regression, the results of these two methods are similar. When ignoring the regulatory role of green energy and green production, green finance negatively impacts green growth. This result is entirely opposite when considering the regulatory role of green energy and green production, green finance impacts green growth positively. However, Bayesian regression is more effective when providing different posterior probability intervals and probability ranges for independent variables to affect the dependent variable. Specifically, the probability that green financial growth has a negative impact on green growth is above 75.86 %. Similarly, under the role of green energy, the probability that green finance growth has a positive impact on green growth is 80.45 % and under the role of green production, this probability is 76.64 %. These findings imply that countries should build a financial system associated with the goal of green energy and green production, thereby helping the economy become greener.

On Ramsey’s Rejection of Keynes’s Mathematical Lattice Structure and Interval Valued Probabilities in the A Treatise on Probability

Henry E. Kyburg was able to demonstrate that Keynes had a qualitative, graphical understanding of interval probability based on his careful analysis of Keynes’s diagram on page 39 of the A Treatise on Probability. Kyburg showed in four different papers, published in 1995,1999,2003 and 2010, that Keynes’s diagram on page 39 is a mathematical lattice structure encompassing interval valued probability. However, Kyburg rejected any conclusion that Keynes had provided a mathematical, quantitative theory of interval valued probability. Kyburg did recognize that Keynes had an intuitive understanding about the nature of interval valued probability, but that the best that Keynes had been able to accomplish in this regard in his A Treatise on Probability was to offer some hints, ideas, or suggestions. This is the same conclusion put forth by all members of ISIPTA since 1999. Of course, Keynes had provided a complete mathematical, quantitative theory, in Parts II and III of the A Treatise on Probability, of interval valued probability in chapters XV, XVI, XVII, XX, and XXII. Keynes’s interval valued theory was based on Boole’s original theory of interval valued probability that Kyburg and all members of ISIPTA have overlooked. Kyburg makes a very interesting point about the deficiencies of Ramsey, as regards Keynes’s graphical presentation (see Keynes, 1921, p.161, ft.2) that were based on Boole’s approach. Boole’s mathematical, lattice structure of upper and lower probabilities, that Boole demonstrated had least and greatest limits, narrowest limit, maximum limit, highest, inferior numerical limit, highest minor limit, greatest minor numerical limit, etc. (Boole,1854,pp.288,293,305- 313,317-324)all involve a mathematical lattice structure .All of these terms mean that Boole is solving for a greatest lower bound and/or a least upper bound with his solutions methods, which involved using second order, quadratic equations and third order ,cubic equations. Such greatest lower bounds and least upper bounds automatically specify a mathematical lattice structure to Boole’s partial orderings of probabilities: “What is curious is that the mathematician -philosopher, Frank Ramsey, paid no attention to this structure in his review of the Treatise (Ramsey, 1922) …” (Kyburg,2010, p.26). Actually, Kyburg must have known that Ramsey had never made any comment on Keynes’s graphical demonstration on page 39 of Keynes’s A Treatise on Probability in any publication in his life. The reason is very simple. Ramsey knew that the mathematical lattice demonstration on p.39 showed that his theory of exact and precise, additive probability, illustrated by Keynes with the linear line OAI, is a very, very special case of Keynes’s general theory. Therefore, Ramsey had to reject the foundation for the construction of any type of mathematical lattice structure, which was Keynes’s Boolean relational, propositional logic. Kyburg recognized that Ramsey’s criticism was directed at this foundation: “That there is an out-and-out conflict between Keynes and Ramsey ….becomes clear in the fourth part of Ramsey’s essay…Now it is all very well for Ramsey modestly to admit that he sees no logical relation of probability such as the one that Keynes seeks to draw our attention to[author’s note-Kyburg has completely overlooked that such an objective , logical probability relation was first specified by Boole, not Keynes],but wants to go further than that. It is clear that Ramsey wants to claim that there is no such relation.” (Kyburg, 2010, p.29; italics is Kyburg’s). Ramsey, in fact, did claim this. I have noted many, many times in the papers that I have written on this topic over the last 15 years that are available at SSRN, Researchgate and Academia, that Keynes, if he had actually ever thought himself threatened by Ramsey’s claims, could have, either with or without Bertrand Russell’s assistance ,completely and totally crushed Ramsey intellectually by simply making the following hypothetical one sentence statement at any Apostles meeting from 1922 to 1929: “My objective, logical probability relations are identical to Boole’s logical, probability relations as discussed in chapter I and XVI of his The Laws of Thought (1854).” That, of course, would have been the beginning of the end of Ramsey’s academic career

Groundwater health risk assessment and its temporal and spatial evolution based on trapezoidal fuzzy number-Monte Carlo stochastic simulation: A case study in western Jilin province

The United States Environmental Protection Agency (USEPA) Four-step-Method (FSM) is a straightforward and extensively utilized tool for evaluating regional health risks, However, the complex and heterogeneous groundwater environment system causes great uncertainty in the assessment process. Triangular stochastic simulation (TSS) possesses certain advantages in solving uncertainty problems, but its inadequacy with discrete data reveals limitations in this aspect. To solve the above problems, this study proposes to construct trapezoidal fuzzy number-Monte Carlo stochastic simulation (TFN-MCSS) to compensate for the shortcomings of the first two methods. This method adopted trapezoidal fuzzy number (TFN) analysis to comprehensively consider the characteristics of a large dispersion of water quality monitoring data and the uncertainty of the human health risk assessment (HHRA) process. Concurrently, to overcome the subjectivity and uncertainty of artificially determining the interval of TFN in traditional methods, the slope was used to select the most probable interval value (TMPIV) of TFN combined with the α-truncated set technique (α-TST) and MCSS. Based on these, a TFN-MCSS was constructed and applied to groundwater HHRA in western Jilin Province. First, the groundwater chemical characteristic determination and water quality evaluation in western Jilin were performed to identify the main pollution indicators, and the health risk effects of pollutants in groundwater of different aquifers at different time periods on adults and children were evaluated using the TFN-MCSS. The uncertainty and sensitivity were analyzed, and the primary risk control indicators were identified and compared to FSM and TSS. The results reveal that TFN-MCSS was more sensitive to data and could reduce the uncertainty of assessment process. It indicated that over a 10-year period, the health risks associated with unconfined groundwater (UW) and confined water (CW) decreased by greater than 52 %. However, the highest total non-carcinogenic risk index (THI) was 1.3-fold higher than the safety threshold, and this posed a health risk.

Form Uncertainty to Sustainable Decision-Making: A Novel MIDAS–AM–DeepAR-Based Prediction Model for E-Commerce Industry Development

Global efforts to build sustainable e-commerce ecosystems through various prediction tools have suffered due to uncertainty in politics, the economy, and the environment. This paper proposes a new integrative prediction model to track the sustainable development of e-commerce. Using US e-commerce data, this study explores the prediction accuracy of the mixed data sampling (MIDAS) model in combination with the attention mechanism (AM) approach, analyzing the performance differences between the model’s before and after improvements. More so, the paper evaluates the performance of the new prediction approach against other competing models using the prediction accuracy metric, the probability interval test, and the Diebold and Mariann (DM) test methods. The results indicate that the introduction of the MIDAS and the AM approaches allows the prediction model to fully utilize the effective information of the mixed-frequency data while simultaneously capturing the differences in the importance of the variables in terms of their time series and the non-linear relationship of the learning variables, thereby positively influencing the economic prediction of the e-commerce industry. Second, the proposed prediction model combines the ability of long-term and short-term high-precision prediction and performs multi-step probability prediction on the development of the e-commerce industry. It can better track abnormal changes in macroeconomic indicators and fit their fluctuation trends. Third, based on the results of the three evaluation indicators, the MIDAS–AM–Deep autoregressive recurrent neural network (DeepAR) model achieves optimal prediction accuracy, allowing it to provide more timely, accurate, and comprehensive predictions for e-commerce management decisions when macroeconomic conditions are undergoing significant transformations.

Research on Calculation of Porosity of Rammed Soil in Pingyao City wall (China) Based on Scanning Electron Microscope Images

ABSTRACT Porosity is an important parameter that affects the deformation, mechanical properties, and constitutive relationships of soil structures. It can be used as an important basis for deterioration research and safety assessment in the field of earth site protection. By collecting scanning electron microscope images (SEM) of the rammed earth of the ancient city wall of Pingyao, a world cultural heritage, and using software such as ImageJ and MATLAB to analyze the characteristics of the images from two-dimensional and three-dimensional perspectives, the relationship between threshold and porosity was studied and summarized. It was found that using the Boltzmann function eigenvalue as the control threshold can well characterize the two-dimensional porosity of soil. Using the idea of integration, the binary slice images with different thresholds are simplified and accumulated to construct a three-dimensional image porosity calculation model. Write a three-dimensional pore characteristic parameter program to extract the statistical image grayscale information and perform normal fitting. Based on the probability interval of the image grayscale distribution, calculate the threshold interval that can effectively reflect the three-dimensional pore characteristics of the soil. This research has important theoretical and practical significance for the protection research of earthen sites.

Conditional bias adjusted estimator of treatment effect in 2-in-1 adaptive design

ABSTRACT For implementation of adaptive design, the adjustment of bias in treatment effect estimation becomes an increasingly important topic in recent years. While adaptive design literature traditionally focuses on the control of type I error rate and the adjustment of overall unconditional bias, the research on adjusting conditional bias has been limited. This paper proposes a conditional bias adjustment estimator of treatment effect under the context of 2-in-1 adaptive design and aims to provide a comprehensive investigation on their statistical properties including bias, mean squared error and coverage probability of confidence intervals. It demonstrated that conditional bias adjusted estimators greatly reduce the conditional bias and have similarly negligible unconditional bias compared with mean and median (unconditional) unbiased estimators. In addition, the test statistics is constructed based on the conditional bias adjustment estimators and compared with the naive unadjusted test.

Fast Bayesian Functional Regression for Non-Gaussian Spatial Data

Functional generalized linear models (FGLM) have been widely used to study the relations between non-Gaussian response and functional covariates. However, most existing works for FGLM assume independence among observations and therefore they are of limited applicability for correlated data. A particularly important example is spatial functional data, where we observe functions over spatial domains, such as the age population curve or temperature curve at each areal unit. In this paper, we extend FGLM by incorporating spatial random effects. Especially, we study the relationship between the non-Gaussian response variable and functional covariates that are spatially observed. However, such model has computational and inferential challenges. The high-dimensional spatial random effects cause the slow mixing of Markov chain Monte Carlo (MCMC) algorithms. Furthermore, spatial confounding can lead to bias in parameter estimates and inflate their variances. To address these issues, we propose an efficient Bayesian method using a sparse reparameterization of high-dimensional random effects. We also study the average coverage probabilities of the credible intervals of functional parameters. We apply our methods to simulated and real data examples, including malaria incidence data and US COVID-19 data. The proposed method is fast while providing accurate functional estimates.

A phase 1/2 study of favezelimab in combination with pembrolizumab for heavily pretreated anti–PD-1–refractory classical Hodgkin lymphoma (cHL): An updated analysis.

7056 Background: PD-1 inhibitors are standard of care for relapsed or refractory (R/R) cHL, but patients (pts) often progress after anti–PD-1 therapy. The immune checkpoint receptor lymphocyte-activation gene 3 (LAG-3) is proposed to contribute to anti–PD-1 resistance. In a phase 1/2 study (NCT03598608) evaluating favezelimab (anti–LAG-3) plus pembrolizumab (anti–PD-1) for R/R hematologic malignancies, the combination provided antitumor activity and manageable safety in pts with heavily pretreated anti–PD-1–refractory cHL (cohort 2). Updated results after additional follow-up are presented. Methods: Eligible pts were aged ≥18 years with R/R cHL; prior treatment with or ineligibility for autologous stem cell transplantation or no response to salvage chemotherapy; an ECOG performance status of 0 or 1; and disease progression after ≥2 doses of anti–PD-1–based therapy and within 12 weeks of the last dose of anti–PD-1 therapy. The study comprised a safety lead-in (pembrolizumab 200 mg IV Q3W + favezelimab 200-mg starting dose, then dose escalation to 800 mg IV Q3W per a modified toxicity probability interval design) and a dose expansion phase (pembrolizumab + favezelimab at the established recommended phase 2 dose of 800 mg Q3W for ≤35 cycles). Primary end point was safety. ORR per IWG 2007 criteria by investigator review was a secondary end point. DOR and PFS per IWG 2007 criteria by investigator review and OS were exploratory. Results: Cohort 2 enrolled 34 pts. Median time from first dose to data cutoff (August 15, 2023) was 40.7 months (range, 20.4-54.8). Treatment-related AEs occurred in 28 pts (82%); the most common (≥15%) were hypothyroidism (18%), nausea (18%), and fatigue (15%). Grade 3 or 4 treatment-related AEs occurred in 6 pts (18%). Discontinuation due to treatment-related AEs occurred in 6 pts (18%). No deaths due to treatment-related AEs were reported. AEs of clinical interest occurred in 17 pts (50%); 2 pts (6%) had grade 3 events (encephalitis, hepatitis) and 1 pt (3%) had a grade 4 event (type 1 diabetes mellitus). ORR was 29% (10/34; 95% CI, 15-48), with 3 (9%) complete responses and 7 (21%) partial responses. ORR was 35% (6/17; 95% CI, 14-62) in pts with anti–PD-1 and 24% (4/17; 95% CI, 7-50) in pts with non–anti–PD-1 as most recent therapy. Median DOR was 21.9 months (range, 0.0+ to 26.1+) and an estimated 17% of responders remained in response at 24 months. Median PFS was 9.7 months (95% CI, 5.1-14.7); 24-month PFS rate was 21%. Median OS was 39.0 months (95% CI, 25.7-not reached); 24-month OS rate was 73%. Conclusions: After additional follow-up, favezelimab + pembrolizumab continued to demonstrate manageable safety and antitumor activity in pts with heavily pretreated anti–PD-1–refractory cHL. The phase 3 KEYFORM-008 study (NCT05508867) will evaluate a coformulation of favezelimab and pembrolizumab in pts with anti–PD-1–refractory cHL. Clinical trial information: NCT03598608 .