Moment Estimation Research Articles

Multi-Objective Optimization Algorithm Based Bidirectional Long Short Term Memory Network Model for Optimum Sizing of Distributed Generators and Shunt Capacitors for Distribution Systems

In this paper, a multi-objective grey wolf optimization (GWO) algorithm based Bidirectional Long Short Term Memory (BiLSTM) network machine learning (ML) model is proposed for finding the optimum sizing of distributed generators (DGs) and shunt capacitors (SHCs) to enhance the performance of distribution systems at any desired load factor. The stochastic traits of evolutionary computing methods necessitate running the algorithm repeatedly to confirm the global optimum. In order to save utility engineers time and effort, this study introduces a BiLSTM network-based machine learning model to directly estimate the optimal values of DGs and SHCs, rather than relying on load flow estimates. At first, a multi-objective grey wolf optimizer determines the most suitable locations and capacities of DGs and SHCs at the unity load factor and the same locations are used to obtain optimum sizing of DGs and SHCs at other load factors also. The base case data sets consisting of substation apparent power, real power load, reactive power load, real power loss, reactive power loss and minimum node voltage at various load factors in per unit values are taken as input training data for the machine learning model. The optimal sizes of the DGs and SHCs for the corresponding load factors obtained using GWO algorithm are taken as target data sets in per unit values for the machine learning model. An adaptive moment estimation (adam) optimization approach is employed to train the BiLSTM ML model for identifying the ideal values of distributed generations and shunt capacitors at different load factors. The efficacy of the proposed ML-based sizing algorithm is demonstrated via simulation studies.

Age of air from in situ trace gas measurements: insights from a new technique

Abstract. The age of air is an important transport diagnostic that can be derived from trace gas measurements and compared to global chemistry climate model output. We describe a new technique to calculate the age of air, measuring transport times from the Earth's surface to any location in the atmosphere based on simultaneous in situ measurements of multiple key long-lived trace gases. The primary benefits of this new technique include (1) optimized ages of air consistent with simultaneously measured SF6 and CO2; (2) age of air from the upper troposphere through the stratosphere; (3) estimates of the second moment of age spectra that have not been well constrained from measurements; and (4) flexibility to be used with measurements across multiple instruments, platforms, and decades. We demonstrate the technique on aircraft and balloon measurements from the 1990s, the last period of extensive stratospheric in situ sampling, and several recent missions from the 2020s, and compare the results with previously published and modeled values.

Novel closed-form point estimators for the beta distribution

In this paper, we propose and investigate novel closed-form point estimators for the beta distribution. The estimators of the first type are a modified version of Pearson's method of moments. The underlying idea is to involve the sufficient statistics, i.e., log-moments in the moment estimation equations and solve the mixed type of moment equations simultaneously. The estimators of the second type are based on an approximation to Fisher's likelihood principle. The idea is to solve two score equations derived from the log-likelihood function of generalized beta distributions. Both two resulted estimators are in closed forms, strongly consistent and asymptotically normal. In addition, through theoretical analyses and extensive simulations, the proposed estimators are shown to perform very close to the maximum likelihood estimators in both small and large samples, and they significantly outperform the method of moment estimators.

Investigating the mediating role of national governance bundles and institutional ownership on the relationship between risk governance disclosure and market valuation: evidence from Sub-Saharan Africa

Purpose The paper aims to comparatively examine the impact of risk governance disclosure (RGD) on the market valuation of firms in Sub-Saharan Africa (SSA) and the mediating role of institutional investment and national governance bundles (NGB). Design/methodology/approach Using a dynamic system generalized method of moments estimation to control for endogeneity, the data for this research is manually collected from the annual reports of small and large firms in Nigeria (80 firms) and South Africa (100 firms) for the period 2012–2017 (900 firm years). Findings The authors find that firm RGD directly impacts firm valuation positively, but this association is significantly mediated by national governance practices (bundles) and institutional investment. The authors also develop a conceptual framework that shows the direct and indirect impact of RGD on firm market valuation. Originality/value The paper contributes to the comparative corporate governance literature in three ways. First, the authors show that differences in country-level RGD are explained by the maturation of governance regulations and institutions in each country. Second, despite the differences in the level of maturity of governance institutions across countries, stock markets value risk governance information. Finally, the study develops a conceptual framework that addresses prior inconsistent findings by showing that firm-level NGB and institutional investment significantly mediate the association between RGD and market valuation.

AdaLo: Adaptive learning rate optimizer with loss for classification

Gradient-based algorithms are frequently used to optimize neural networks, with various methods developed to enhance their performance. Among them, the adaptive moment estimation (Adam) optimizer is well-known for its effectiveness and ease of implementation. However, it suffers from poor generalization without a learning rate scheduler. Additionally, it has the disadvantage of a large computational burden because of individual learning rate term, as known as second-order moments of gradients. In this study, we propose a novel gradient descent algorithm called AdaLo, which stands for Adaptive Learning Rate Optimizer with Loss. AdaLo addresses two problems using its adaptive learning rate (ALR). Firstly, the proposed ALR adjusts the learning rate, based on the model's training progress, specifically the loss value. Therefore AdaLo's ALR effectively replaces traditional learning rate schedulers. Secondly, the ALR is a scalar global learning rate, reducing the computational burden. In addition, the stability of the proposed method is analyzed from the perspective of the learning rate. The superiority of AdaLo was proven by non-convex functions. Simulation results indicated that the proposed optimizer outperformed the Adam, AdaBelief, and diffGrad with regard to the training error and test accuracy.

Reweighted Nadaraya–Watson estimation of stochastic volatility jump-diffusion models

In this paper, we construct the reweighted Nadaraya–Watson estimators of the infinitesimal moments for the volatility process of the stochastic volatility models, with the application of the threshold estimator of the unobserved volatility process. Our model includes jumps in both the underlying asset price and its volatility process. We derive the asymptotic properties of the estimators under the infill and long span assumptions. The results are useful for identification of the process. The finite-sample performance of the estimators is studied through Monte Carlo simulation.

On the estimation of the switching moment of utility functions in cooperative differential games

On the estimation of the switching moment of utility functions in cooperative differential games

Prediction of fracture toughness of concrete using the machine learning approach

In the process of structural design, it is useful to estimate the fracture toughness of concrete samples. This research showcases the effectiveness of utilizing machine learning methods to determine the fracture toughness of concrete. Taking into account variables such as mix design, machine learning techniques can accurately predict the mode I fracture toughness of concrete. Dimensionless stress intensity factor of concrete prediction using twelve different machine learning techniques namely, Linear regression (LR), Extreme Gradient-Boosting (XGboost), K-Nearest Neighbors (KNN), Random Forest (RF), Category Boosting (CB), Decision Tree (DT), Extra Trees (ET), Light Gradient-Boosting (LightGB), Adaptive boosting (AdaBoost), Bagging (BA), Gaussian Process (GP), Artificial Neural Network (ANN) and Support Vector Machine (SVM). The result of utilizing the training adaptive moment estimation algorithm has been developed to create an outstanding machine learning-based system. After carefully analyzing comparisons between the predictions produced by different models and experimental findings, it has been discovered that the models demonstrate an impressive accuracy rate of about 90 percent when it comes to forecasting concrete fracture toughness. The research findings emphasize that the ANN model exhibited superior accuracy in its predictions (R2 value of 0.90, RMSE of 0.1517, and MAE of 0.1238). Upon conducting a thorough examination of the ANN method’s sensitivity, the cement parameter holds utmost significance in accurately estimating concrete’s fracture toughness using the available dataset. So, the ANN model can be used as a valuable method to provide practical assistance in predicting the fracture toughness of concrete. When evaluating the effective parameters, the cement and metakaolin dosage and the notch height to specimen height ratio have the greatest effect on the fracture resistance, while the coarse aggregate content is minimal. This result is consistent with the experimental data.

How Does Tax Avoidance Affect a Firm's Investment Inefficiency in Taiwan's Semiconductor Industry?

ABSTRACTThis study examines an important issue regarding how tax avoidance affects a firm's investment inefficiency in Taiwan's semiconductor industry. In addition, the dynamic effect of a firm's investment inefficiency and spillover effect of investment inefficiency from competitors are also explored. This study adopts 44 listed firms from 2013 to 2021 and constructs the “competitor‐dependence matrix” to replace the “spatial‐dependence matrix” in the dynamic spatial autoregressive model. According to the difference generalized method of moments estimation, the primary finding of this study is that the tax avoidance has a statistically and significantly positive effect on a firm's investment inefficiency while it is overinvestment, but a negative effect while it is underinvestment. Moreover, the dynamic effect is statistically insignificant, but the spillover effect is statistically and significantly negative.

The impact of information and communication technologies on export diversification: Evidence from developing countries

As economies increasingly rely on digitalization, information and communication technologies (ICT) have become critical in shaping global economic transformations. This study explores the impact of ICT on export concentration and diversification in emerging markets and developing countries. Using a panel data set of 110 countries from 2000 to 2019, the Generalized Method of Moments (GMM) estimator has been employed to assess the short- and long-term effects of ICT on these economies. To summarize information on various ICT channels, an ICT index has been constructed through principal component analysis (PCA). Additionally, the study examines the combined influence of ICT and human capital on export concentration. Findings indicate that ICT development significantly accelerates export diversification in developing countries, aligning their export structures with global standards. The interaction between ICT and human capital positively impacts export concentration, suggesting that higher levels of human capital enhance ICT's effects. Overall, ICT not only boosts human capital but also equips developing countries with the skills needed to enter new markets, develop high-value products, and compete globally. Consequently, governments in developing countries should prioritize investments in ICT infrastructure to lower trade costs, diversify exports, and promote sustainable economic growth.

Efficient Fault Detection and Analysis of Power System Distribution Networks by Integrating BP Data Mining

As the basic guarantee for people’s production and life, the safe operation of the power system has an important impact on the development and operation of society. To ensure the safe and stable operation of the power grid, predicting potential faults and taking reasonable preventive measures can effectively avoid the occurrence of power accidents. However, due to the difficulty in ensuring the prediction accuracy of traditional methods, there are issues of protection misoperation and rejection. Therefore, in order to achieve accurate prediction of power grid faults and avoid protection misoperation and rejection issues, a distribution network fault classification prediction model using a combination of three-layer data mining model (TLDM) and adaptive moment estimation (Adam) algorithm/random gradient descent algorithm improved backpropagation neural network (BPNN) is proposed. The implementation results showed that the classification accuracy of artificial fish school [Formula: see text] priori, [Formula: see text]-means clustering convolutional neural network model and TLDM for single-phase grounding faults was 93.2%, 91.5% and 96.6%, respectively. The classification accuracy for two-phase faults was 92.8%, 92.4% and 95.7%, respectively. The classification accuracy for two-phase grounding faults was 93.7%, 91.2% and 96.9%, respectively. The classification accuracy for three-phase faults was 93.3%, 92.1% and 97.1%, respectively. The TLDM had the highest classification accuracy. The average accuracy, average accuracy and average recall of the BPNN improved by the combination of the ADAM algorithm and random gradient descent algorithm were 94.1%, 90.9% and 88%, respectively, which were higher than the BPNN improved by the combination of ADAM algorithm and random gradient descent algorithm. The above results indicate that the proposed distribution network fault classification and prediction model has good performance and can achieve accurate prediction of distribution network faults.

Group Sparse β -Model for Network

Sparsity, homogeneity and heterogeneity are three important characteristics of many real-life networks. The recently proposed Sparse β -Model divides nodes into core ones and peripheral ones to accommodate sparsity, but the parameters of core nodes are assumed to be of similar magnitude, which may not be in line with applications. In this paper, we propose the Group Sparse β -Model that splits the core nodes into groups and assumes different orders of magnitude of parameters in different groups, accounting for the heterogeneity among core nodes. When the groups are known, we provide consistent and asymptotically normal moment estimators of the parameters that control the global and local density. Based on that, consistency and asymptotic normality of the maximum likelihood estimators of the remaining parameters are derived. We also establish finite-sample error bounds results. When the groups are unknown, a ratio method is proposed to detect groups, which is computationally efficient. Simulations show competitive results and the analysis of a corporate inter-relationships network illustrates the usefulness of the proposed model.

Telescope Alignment Method Using a Modified Stochastic Parallel Gradient Descent Algorithm

To satisfy the demands of high image quality and resolutions, telescope alignment is indispensable. In this paper, a wavefront sensorless method based on a modified stochastic parallel gradient descent algorithm (SPGD) called the adaptive moment estimation SPGD (Adam SPGD) algorithm is proposed. Simulations are carried out using a four-mirror telescope, whose aperture is 6 m and fields of view are Φ2°. Three misalignments are shown as examples. Positions of the secondary mirror and third mirror are employed to compensate aberrations. The results show that merit functions and energy distributions of corrected images match with the designed ones. The mean RMS of residual wavefront errors is smaller than λ/14 (λ = 0.5 μm), indicating that the misalignments are well compensated. The results verify the effectiveness of our method.

Aspect sensitivity measurement of backscattering radar echoes from 205 MHz Stratosphere–Troposphere radar at a tropical coastal station

Aspect sensitivity in VHF radar refers to the extent to which the power and spectrum width of echoes vary with changes in the zenith angle, which is related to the backscattering process. The scattering of clear air signals is primarily caused by Fresnel reflection, anisotropic scattering, and isotropic scattering. The aspect sensitivity and accuracy of moment and wind estimation in clear air radars are influenced by the beam width, beam pointing angle from zenith, and atmospheric conditions. This study proposes a method to determine the sensitivity of the recently developed Stratosphere–Troposphere (ST) wind profile Radar in Cochin, India (10.04°N, 76.33°E). The radar operates at a distinct frequency of 205 MHz in the far VHF band. The experiment is conducted at an altitude of 5 to 20 km above the ground by adjusting the beam’s orientation with a resolution of 2°in both the east–west and north-south directions. The estimation of wind components is subject to uncertainty due to the varying aspect angles caused by the distinct dispersion properties in this height range. The study found that power variance is lowest between 6 and 12 km in both north-south and east–west directions, while daily fluctuations in aspect-sensitive echoes complicate wind component estimation. Correlation length (ζ) ranges from 0.5 to 15 m, indicating various air scattering processes. Notably, θs is smaller near the zenith and increases with tilt angles, exceeding 20°up to 14 km before declining at higher altitudes, indicating significant anisotropy at elevated levels. The wide range of R factor values (0.1 to 0.9) across different heights causes significant ambiguity in wind estimation. In this study, the impact of various aspect sensitivity parameters on wind estimation on days with clear air has been analyzed.

Real-world analysis of the correlation between overall survival and progression-free survival in advanced pancreatic cancer: Results of NAPOLEON-1 and 2 studies.

Fluorouracil, leucovorin, irinotecan, and oxaliplatin (FOLFIRINOX) improve overall survival (OS) and progression-free survival (PFS) in patients with pancreatic cancer, compared with gemcitabine (GEM). However, whether PFS is a surrogate marker of OS in pancreatic cancer chemotherapy focusing on FOLFIRINOX or GEM plus nab-paclitaxel remains unknown. We aimed to verify whether PFS can be a surrogate marker of OS in prognosis prediction. This was an integrated analysis of the NAPOLEON study and retrospective cohort of the NAPOLEON-2 study-a multicenter observational study conducted in Japan, using real-world data. The primary and secondary endpoints were OS and PFS, respectively. The correlation between OS and PFS in first- and second-line treatments was assessed using Method of Moments estimation. An analysis was performed in patients with confirmed OS at the end of follow-up. The NAPOLEON-2 cohort included only patients who received 5-fluorouracil, leucovorin, and nanoliposomal irinotecan (NFF) as second-line treatment. Among 479 patients, the correlation between PFS and OS from first- and second-line chemotherapies was calculated in 310 and 225 patients, respectively. The R-squared values for the correlation between PFS and OS from first- and second-line chemotherapies were 0.74 and 0.76, respectively. There was no statistically significant difference in first-line treatment between the FOLFIRINOX and gemcitabine plus nab-paclitaxel groups (P=0.92). Therefore, the FOLFIRINOX group may not have shown a stronger correlation than the NFF group. PFS can be a surrogate marker of OS in first- and second-line therapies. Appropriate prognostic estimation might contribute to proper treatment selection.

Estimating Social Effects with Randomized and Observational Network Data

Abstract This paper introduces an innovative approach to identifying and estimating the parameters of interest in the widely recognized linear-in-means regression model under conditions where the initial randomization of peers determines the observed network. We assert that peers who are initially randomized do not produce social effects. However, after randomization, agents can endogenously develop significant connections that potentially generate peer influences. We present a moment condition that compiles local heterogeneous identifying information for all agents within the population. Under the assumption of ψ-dependence in the endogenous network space, we propose a Generalized Method of Moments (GMM) estimator, which is proven to be consistent, asymptotically normally distributed, and straightforward to implement using commonly available statistical software due to its closed-form expression. Monte Carlo simulations demonstrate the GMM estimator’s strong small-sample performance. An empirical analysis utilizing data from Hong Kong high school students reveals substantial positive spillover effects on math test scores among study partners in our sample, provided that their seatmates were exogenously assigned by their teachers.

Climate change and economic growth: Evidence for European countries

AbstractClimate change may affect economies and the welfare of people around the world. To design appropriate policy responses, the economic effects of climate change should be known. One strand in the literature empirically estimates the growth effects of climatic variations. However, those studies often neglect economic variables that have proven to be robust in explaining economic growth. Further, often they fail to check for the robustness of their results. The main aim of this study is to detect whether there exists a statistically significant robust relation between climate change and economic growth by estimating different model specifications. To do so panel estimation techniques for 24 European economies for the period from 2002 to 2019 are applied whereby panel fixed effects estimations and dynamic Generalized Methods of Moments estimations are resorted to. No statistically significant robust relationship between the temperature change and economic growth is found just as for precipitation that does not exert a significant effect on growth. As regards the institutional and macroeconomic control variables the rule of law, the fiscal variable, and the output gap are statistically significant and robust. It is argued that far‐reaching policy measures, such as the net zero goal of the European Union, should be given only based on robust results. Otherwise, economic policy may turn out to be inadequate and can lead to welfare losses. Hence, the conclusion is that the net zero goal of the European Green Deal is to be seen skeptical.

Evaluation of Four Forensic Investigative Genetic Genealogy Analysis Approaches with Decreased Numbers of SNPs and Increased Genotyping Errors

Background: Forensic investigative genetic genealogy (FIGG) has developed rapidly in recent years and is considered a novel tool for crime investigation. However, crime scene samples are often of low quality and quantity and are challenging to analyze. Deciding which approach should be used for kinship inference in forensic practice remains a troubling problem for investigators. Methods: In this study, we selected four popular approaches—KING, IBS, TRUFFLE, and GERMLINE—comprising one method of moment (MoM) estimator and three identical by descent (IBD) segment-based tools and compared their performance at varying numbers of SNPs and levels of genotyping errors using both simulated and real family data. We also explored the possibility of making robust kinship inferences for samples with ultra-high genotyping errors by integrating MoM and the IBD segment-based methods. Results: The results showed that decreasing the number of SNPs had little effect on kinship inference when no fewer than 164 K SNPs were used for all four approaches. However, as the number decreased further, decreased efficiency was observed for the three IBD segment-based methods. Genotyping errors also had a significant effect on kinship inference, especially when they exceeded 1%. In contrast, MoM was much more robust to genotyping errors. Furthermore, the combination of the MoM and the IBD segment-based methods showed a higher overall accuracy, indicating its potential to improve the tolerance to genotyping errors. Conclusions: In conclusion, this study shows that different approaches have unique characteristics and should be selected for different scenarios. More importantly, the integration of the MoM and the IBD segment-based methods can improve the robustness of kinship inference and has great potential for applications in forensic practice.

Automated lepidopteran pest developmental stages classification via transfer learning framework.

The maize crop is highly susceptible to damage caused by its primary pests, which poses considerable challenges in manually identifying and controlling them at various larval developmental stages. To mitigate this issue, we propose an automated classification system aimed at identifying the different larval developmental stages of 23 instars of 4 major lepidopteran pests: the Asian corn borer, Ostrinia furnacalis (Guenée; Lepidoptera: Crambidae), the fall armyworm, Spodoptera frugiperda (J.E. Smith; Lepidoptera: Noctuidae), the oriental armyworm, Mythimna separata (Walker; Lepidoptera: Noctuidae), and the tobacco cutworm, Spodoptera litura (Fabricius; Lepidoptera: Noctuidae). Employing 5 distinct Convolutional Neural Network architectures-Convnext, Densenet121, Efficientnetv2, Mobilenet, and Resnet-we aimed to automate the process of identifying these larval developmental stages. Each model underwent fine-tuning using 2 different optimizers: stochastic gradient descent with momentum and adaptive moment estimation (Adam). Among the array of models tested, Densenet121, coupled with the Adam optimizer, exhibited the highest classification accuracy, achieving an impressive 96.65%. The configuration performed well in identifying the larval development stages of all 4 pests, with precision, recall, and F1 score evaluation indicators reaching 98.71%, 98.66%, and 98.66%, respectively. Notably, the model was ultimately tested in a natural field environment, demonstrating that Adam_Densenet121 model achieved an accuracy of 90% in identifying the 23 instars of the 4 pests. The application of transfer learning methodology showcased its effectiveness in automating the identification of larval developmental stages, underscoring promising implications for precision-integrated pest management strategies in agriculture.

Uncertain Logistic Population Model with Allee Effect

By considering the influence of various uncertain noises, we established an uncertain Logistic population model with Allee effect, which was characterized by an uncertain differential equation. Firstly, we obtained the solution of the model, and discussed the stability of the equilibrium state. Secondly, the unknown parameters in the model were estimated by using the generalized moment estimation method under the framework of uncertainty theory. Finally, the parameter estimations of the model and the properties of the solution were explained by an example analysis.