Moment Estimation Research Articles

Single Stage Shrinkage Estimation Methods for Reliability Function of Generalized Exponential Distribution Using Simulation

The reliability in stress–strength model (s–s) is estimated in this article when the system has parallel and series components subject to one of the stresses and it follows generalized exponential distribution by taking via different estimation methods, namely the maximum likelihood estimation, unbiased estimation, moment estimation and shrinkage estimation. Also, the Mont Carlo method is used to compare these methods under mean squared error.

Market context matters: the impact of corporate diversification on REIT performance in Malaysia and Singapore

PurposeThis study aims to investigate the impact of corporate diversification and growth opportunities on the performance of real estate investment trusts (REIT) in Malaysia and Singapore before and during the pandemic.Design/methodology/approachThe sample consists of 33 public-listed REITs across Singapore and Malaysia. A dynamic panel system generalized method of moments (DPS-GMM) estimation is used to account for unobservable factors and a relatively short sample period (2009–2022).FindingsResults indicate that the impact of diversification is contingent on the market where the REIT is based and other institutional factors. The estimates also show that diversified REITs are better able to weather period of economic uncertainty.Practical implicationsWe provided a definitive answer as to why corporate diversification leads to conflicting outcomes – market and institutional factors, strategic intent and the overall economic environment. We also show that the impact of typical firm controls (i.e. free cash, size) can differ. Future firm-level work should thus study similar phenomenon more contextually and carefully consider these varying effects.Originality/valueThe literature is divided on the impact of diversification on firm performance. By using a two-country sample, we show conclusive evidence that this contradictory outcome is due to market and institutional factors. We also show evidence that strategic intent is an important factor that influences the outcomes of diversification, regardless of market. We also infer that excess cash aids the resilience of the firm, contrary to the negative perception of excess cash during normal times. Firm size, in contrast, does not contribute to firm performance during a crisis.

A novel correction method for modelling parameter-driven autocorrelated time series with count outcome

BackgroundCount time series (e.g., daily deaths) are a very common type of data in environmental health research. The series is generally autocorrelated, while the widely used generalized linear model is based on the assumption of independent outcomes. None of the existing methods for modelling parameter-driven count time series can obtain consistent and reliable standard error of parameter estimates, causing potential inflation of type I error rate.MethodsWe proposed a new maximum significant ρ correction (MSRC) method that utilizes information of significant autocorrelation coefficient ρ estimate within 5 orders by moment estimation. A Monte Carlo simulation was conducted to evaluate and compare the finite sample performance of the MSRC and classical unbiased correction (UB-corrected) method. We demonstrated a real-data analysis for assessing the effect of drunk driving regulations on the incidence of road traffic injuries (RTIs) using MSRC in Shenzhen, China. Moreover, there is no previous paper assessing the time-varying intervention effect and considering autocorrelation based on daily data of RTIs.ResultsBoth methods had a small bias in the regression coefficients. The autocorrelation coefficient estimated by UB-corrected is slightly underestimated at high autocorrelation (≥ 0.6), leading to the inflation of the type I error rate. The new method well controlled the type I error rate when the sample size reached 340. Moreover, the power of MSRC increased with increasing sample size and effect size and decreasing nuisance parameters, and it approached UB-corrected when ρ was small (≤ 0.4), but became more reliable as autocorrelation increased further. The daily data of RTIs exhibited significant autocorrelation after controlling for potential confounding, and therefore the MSRC was preferable to the UB-corrected. The intervention contributed to a decrease in the incidence of RTIs by 8.34% (95% CI, -5.69–20.51%), 45.07% (95% CI, 25.86–59.30%) and 42.94% (95% CI, 9.56–64.00%) at 1, 3 and 5 years after the implementation of the intervention, respectively.ConclusionsThe proposed MSRC method provides a reliable and consistent approach for modelling parameter-driven time series with autocorrelated count data. It offers improved estimation compared to existing methods. The strict drunk driving regulations can reduce the risk of RTIs.

Implementation of Neural Network in Early Detection of Financial Crisis in Singapore

The financial crisis that occurred in 1997 and 2008 had a negative impact on several countries, including Singapore. A financial crisis can occur suddenly so it can endanger a country’s economy if it is not prepared for it. Therefore, early detection of financial crises is needed as a form of crisis warning so that the government can anticipate and prepare appropriate policies. The independent variables used are monthly data of 11 key macroeconomic and financial indicators of Singapore’s economy from January 1990 to June 2021. The Perfect signal is used as the dependent variable in the crisis early detection system. This study aims to build a model of a financial crisis detection system in Singapore using Multilayer Perceptron Backpropagation (MLPBP) as a neural network algorithm by comparing the optimization of Stochastic Gradient Descent (SGD) and Nesterov-accelerated Adaptive Moment Estimation (Nadam). The optimal hyperparameter value in the model was searched using the grid search method based on the accuracy and obtained the best model with 11-11-1 network architecture, best optimization is Nadam, learning rate = 0.1;

Unravelling open innovation determinants and firm performance relationships in CEE and SEE countries

Purpose Open innovation (OI), by now one of the major concepts for the analysis of innovation, is seen as a methodology for collaboratively designing and implementing solutions by engaging stakeholders in an iterative and inclusive service design process. This paper aims to empirically investigate OI capacities, defined as a cooperative, knowledge-sharing innovation ecosystem, and to explore how it can lead to improved performance of firms in Central and Eastern European (CEE) and Southeastern European (SEE) countries.Design/methodology/approach The study builds on the World Bank/European Bank for Reconstruction and Development (EBRD’s) Business Environment Enterprise Performance Survey (BEEPS) dataset for 2009, 2013 and 2019. Primarily, the research model was estimated using log-transformed ordinary least squares (OLS). Taking into consideration that this method might produce substantial bias, yielding misleading inferences, this study is fitting Poisson pseudo maximum likelihood estimators with robust standard errors and instrumental variable/generalized method of moments estimation (IV/GMM) approach for comparative results. Secondarily, the research model was tested using structural equation modelling (SEM) to investigate the relationship between five OI capacities and firm performance.Findings The findings indicate that there is a significant positive relationship between most OI capacities and firm performance, except for innovation, which did not show a statistically significant relationship with firm performance. Specifically, research and development (R&D), knowledge and coopetition are statistically significant and positively associated with firm performance, whereas transformation is statistically significant but negatively associated with firm performance. The IV/GMM estimations’ findings support the view that the firm performance is significantly affected by OI capacities, together with some control variables such as size, age, foreign ownership and year dummy to have a significant impact on firm performance.Originality/value This paper fills an identified gap in the literature by investigating the impact of OI on firm performance executed in the specific CEE and SEE country context.

Production- and consumption-based emissions in carbon exporters and importers: A large panel data analysis for the EKC hypothesis

The Environmental Kuznets Curve (EKC) hypothesis has become a widely used framework for the investigation of the income-environment nexus. However, the literature in this field is biased toward the reliance on production-based carbon dioxide (CO2) emissions (PBE) as a proxy for environmental degradation. This production perspective ignores the carbon emissions embedded in trade, which are estimated to make up approximately a quarter of total global emissions. Therefore, the carbon leakage issue makes consumption-based CO2 emissions (CBE) accounting vital for better understanding the relationship between income and carbon emissions. In this study, we comparatively investigate the EKC pattern based on both PBE and CBE for a sample of 85 countries over the period between 1990 and 2020. In doing so, we investigate whether different emission accounting methods have an impact on the EKC pattern and examine how this impact differs depending on whether the countries are net importers or exporters of emissions embedded in trade. To this end, we first endogenously classify countries based on the club convergence approach proposed by Phillips & Sul [1] and obtain country groups such as those that are net exporters or importers of CO2 emissions. Using the system generalized method of moments (SGMM) estimator, we test the validity of the EKC hypothesis in the second step and identify the main driving forces behind the changes in PBE and CBE for each convergent club by controlling the effect of trade and urbanization. Therefore, for the first time, we compare the EKC hypothesis based on both PBE and CBE for a large sample of countries that are endogenously divided into different groups, and investigate the potential impact of alternative emission accounting approaches and the status of being net exporters or importers of CO2 emissions on the EKC pattern. The findings support the different patterns for PBE and CBE: while the EKC hypothesis is supported for both PBE and CBE for countries that are net importers of emissions, neither PBE nor CBE show an inverted U-shaped relationship with income for countries that are net exporters of emissions. These results strongly support the need for coordination in policies to reduce CO2 emissions between importers and exporters of hydrocarbons and energy-intensive goods.

Enhancement of Classifier Performance with Adam and RanAdam Hyper-Parameter Tuning for Lung Cancer Detection from Microarray Data-In Pursuit of Precision.

Microarray gene expression analysis is a powerful technique used in cancer classification and research to identify and understand gene expression patterns that can differentiate between different cancer types, subtypes, and stages. However, microarray databases are highly redundant, inherently nonlinear, and noisy. Therefore, extracting meaningful information from such a huge database is a challenging one. The paper adopts the Fast Fourier Transform (FFT) and Mixture Model (MM) for dimensionality reduction and utilises the Dragonfly optimisation algorithm as the feature selection technique. The classifiers employed in this research are Nonlinear Regression, Naïve Bayes, Decision Tree, Random Forest and SVM (RBF). The classifiers' performances are analysed with and without feature selection methods. Finally, Adaptive Moment Estimation (Adam) and Random Adaptive Moment Estimation (RanAdam) hyper-parameter tuning techniques are used as improvisation techniques for classifiers. The SVM (RBF) classifier with the Fast Fourier Transform Dimensionality Reduction method and Dragonfly feature selection achieved the highest accuracy of 98.343% with RanAdam hyper-parameter tuning compared to other classifiers.

Examining the impact of revenue diversity on government productivity using stochastic frontier production function: the U.S. State government

ABSTRACT This study empirically examines the impact of revenue diversity on government productivity across U.S. states via a two-step system generalized method of moments (GMM) estimation during a 41-year period (1980–2020). Government productivity is driven by revenue diversity, economic growth, fiscal health, and population density. In particular, increasing revenue diversity promotes state governments’ productivity, supporting our hypothesis. This result holds even after the mortgage crisis. These new findings suggest that states can enhance their productivity by expanding their revenue sources. Economic growth and rising fiscal health foster government productivity; these results apply after the mortgage crisis as well. An increase in population density also boosts government productivity during the study period. Climbing tax burdens undermine productivity, whereas growth in long-term debt facilitates productivity after the mortgage crisis. This study defines and expands the concept of productivity in U.S. state governments and explores influencing factors. Subsequent research can build on this theoretical model by integrating other variables tied to government productivity.

Estimating the impact of container port throughput on employment: an analysis for African countries with seaports

Ports play a significant role in facilitating international trade and economic development, serving as vital gateways for the movement of goods across the continent and beyond. As global trade volumes continue to rise, efficient port operations hold the potential to not only enhance economic growth but also contribute significantly to job creation across various sectors of the economy. This paper examines the impact of container port throughput on employment in Africa and further tests whether causality runs from employment to container port throughput. To do so, we use a sample of 27 African countries with seaport and data spanning the period from 2010 to 2020 for the analysis. The system- Generalized Method of Moments (SGMM) estimation technique is used as the estimation strategy. We use service, industrial, and total employment percentages of the total population as proxies for employment while annual container throughput measured in Twenty foots Equivalent Units (TEUs) is used as an indicator for port throughput. Based on the empirical results, we establish a positive significant effect of port throughput on employment in Africa. We further show that bidirectional causality exists between port throughput and employment in Africa. Following these findings, we recommend policies that increase port throughput in Africa.

The influence of board policy setting on firm performance in Malaysia: the interacting effect of capital structure

PurposeThe literature has dealt with the relationship between board characteristics (BC) and firm performance (FP) on a large scale. However, it yielded inconsistent results. Thus, this paper aims to examine the indirect relationship between BC and FP through the mediating role of the capital structure (CS).Design/methodology/approachThis study used a sample of 528 non-financial companies listed on Bursa Malaysia from 2015 to 2019. Also, a two-step system generalised method of moments estimation technique was applied.FindingsThe results show that board diversity and the frequency of board meetings positively affect financial performance, and it is negatively influenced by board turnover, size and independence. Also, the results indicate a positive relationship between the independence of the board and all CS variables. Importantly, the findings support the policy-setting role of the board of directors where CS (measured by total debt and short-term debt) suppresses some governance mechanisms’ detrimental effect on FP. Hence, the board of directors, apart from the monitoring function, introduce various policies (financial and non-financial) that enhance the overall performance of companies.Originality/valueThese results are consistent with the agency’s perspective that management practices in selecting the optimal capital reduce agency costs and improve performance. The findings contribute to developing a broader theoretical framework that accounts for the policy-setting role of the board of directors. The current study model of corporate governance offers insight for policymakers into the role of corporate governance other than monitoring functions in organisations and how CS should be taken into consideration with corporate governance and FP association.

DP-AdamBC: Your DP-Adam Is Actually DP-SGD (Unless You Apply Bias Correction)

The Adam optimizer is a popular choice in contemporary deep learning due to its strong empirical performance. However we observe that in privacy sensitive scenarios, the traditional use of Differential Privacy (DP) with the Adam optimizer leads to sub-optimal performance on several tasks. We find that this performance degradation is due to a DP bias in Adam's second moment estimator, introduced by the addition of independent noise in the gradient computation to enforce DP guarantees. This DP bias leads to a different scaling for low variance parameter updates, that is inconsistent with the behavior of non-private Adam, and Adam's sign descent interpretation. We propose the DP-AdamBC optimization algorithm, which corrects for the bias in the second moment estimation and retrieves the expected behaviour of Adam. Empirically, DP-AdamBC significantly improves the optimization performance of DP-Adam by up to 3.5% in final accuracy in image, text, and graph node classification tasks.

Exploiting Geometry for Treatment Effect Estimation via Optimal Transport

Estimating treatment effects from observational data suffers from the issue of confounding bias, which is induced by the imbalanced confounder distributions between the treated and control groups. As an effective approach, re-weighting learns a group of sample weights to balance the confounder distributions. Existing methods of re-weighting highly rely on a propensity score model or moment alignment. However, for complex real-world data, it is difficult to obtain an accurate propensity score prediction. Although moment alignment is free of learning a propensity score model, accurate estimation for high-order moments is computationally difficult and still remains an open challenge, and first and second-order moments are insufficient to align the distributions and easy to be misled by outliers. In this paper, we exploit geometry to capture the intrinsic structure involved in data for balancing the confounder distributions, so that confounding bias can be reduced even with outliers. To achieve this, we construct a connection between treatment effect estimation and optimal transport, a powerful tool to capture geometric information. After that, we propose an optimal transport model to learn sample weights by extracting geometry from confounders, in which geometric information between groups and within groups is leveraged for better confounder balancing. A projected mirror descent algorithm is employed to solve the derived optimization problem. Experimental studies on both synthetic and real-world datasets demonstrate the effectiveness of our proposed method.

The heterogeneous impact of social distancing on household-level economic performances: evidence from a large survey in Vietnam

ABSTRACT This study utilizes a large dataset surveyed by Facebook and several academic institutions to examine the heterogeneous impact of social distancing on household-level economic performances, employing the generalized method of moments (GMM) estimator and controlling multi-channel fixed effects. In particular, in this case study, the Vietnamese government’s sudden social distancing policy, which is seen as a quasi-experiment, has created favourable conditions to evaluate the causality of this nexus. The results confirm that increased social distancing leads to a higher likelihood of unemployment, especially for women and lower education levels groups. Notably, the study also finds that people with less desirable economic status (i.e. higher rate of unemployment and food insecurity) are more heavily hurt by social distancing practices. Therefore, the findings suggest the need for welfare programmes for these disadvantaged groups post-pandemic.

Which banks’ business operations are more risky? The impact of leverage on Brazilian financial institutions

We analyse the effect of leverage components and banks’ ‘business model channels’ on risk-taking for Brazilian banks. Using a detailed dataset from the Brazilian financial system, the results show that the measurement of leverage components is relevant for determining banks’ risk. We highlight two important findings. First, we provide novel insights for the design of a generalised method of moments (GMM) estimation using external instruments. Second, the derivative and on-balance-sheet operations are risky for investment and foreign banks. These results provide important information for banking supervision and regulation and bring new insights regarding banks’ leverage behaviour.

Learning rate burst for superior SGDM and AdamW integration

Current mainstream deep learning optimization algorithms can be classified into two categories: non-adaptive optimization algorithms, such as Stochastic Gradient Descent with Momentum (SGDM), and adaptive optimization algorithms, like Adaptive Moment Estimation with Weight Decay (AdamW). Adaptive optimization algorithms for many deep neural network models typically enable faster initial training, whereas non-adaptive optimization algorithms often yield better final convergence. Our proposed Adaptive Learning Rate Burst (Adaburst) algorithm seeks to combine the strengths of both categories. The update mechanism of Adaburst incorporates elements from AdamW and SGDM, ensuring a seamless transition between the two. Adaburst modifies the learning rate of the SGDM algorithm based on a cosine learning rate schedule, particularly when the algorithm encounters an update bottleneck, which is called learning rate burst. This approach helps the model to escape current local optima more effectively. The results of the Adaburst experiment underscore its enhanced performance in image classification and generation tasks when compared with alternative approaches, characterized by expedited convergence and elevated accuracy. Notably, on the MNIST, CIFAR-10, and CIFAR-100 datasets, Adaburst attained accuracies that matched or exceeded those achieved by SGDM. Furthermore, in training diffusion models on the DeepFashion dataset, Adaburst achieved convergence in fewer epochs than a meticulously calibrated AdamW optimizer while avoiding abrupt blurring or other training instabilities. Adaburst augmented the final training set accuracy on the MNIST, CIFAR-10, and CIFAR-100 datasets by 0.02%, 0.41%, and 4.18%, respectively. In addition, the generative model trained on the DeepFashion dataset demonstrated a 4.62-point improvement in the Frechet Inception Distance (FID) score, a metric for assessing generative model quality. Consequently, this evidence suggests that Adaburst introduces an innovative optimization algorithm that simultaneously updates AdamW and SGDM and incorporates a learning rate burst mechanism. This mechanism significantly enhances deep neural networks’ training speed and convergence accuracy.

Research on magnetic detection target recognition method based on residual network combined with magnetic moment estimation

The unexploded ordnance left over from war conflicts and live-fire exercises has a strong invisibility, posing a serious threat to the safety of local residents' lives and property. Based on the idea of removing useless parameters and regressing the main parameters, we decompose the magnetization field of the dipole target into the product of the basis function and the simplified function, and invert the magnetic moment of the dipole. A method that can quickly estimate the magnetic moment of the detection target, and a detection target recognition method using residual network combined with fast magnetic moment estimation is proposed. Through field experiments for comparison and analysis, the Accuracy of this method for detecting three types of targets can reach 89.7%, Recall can reach 80%, and Precision can reach 90.3%, which provides a technical reference for target type recognition of unexploded ordnance based on magnetic detection.

Estimating human joint moments unifies exoskeleton control, reducing user effort.

Robotic lower-limb exoskeletons can augment human mobility, but current systems require extensive, context-specific considerations, limiting their real-world viability. Here, we present a unified exoskeleton control framework that autonomously adapts assistance on the basis of instantaneous user joint moment estimates from a temporal convolutional network (TCN). When deployed on our hip exoskeleton, the TCN achieved an average root mean square error of 0.142 newton-meters per kilogram across 35 ambulatory conditions without any user-specific calibration. Further, the unified controller significantly reduced user metabolic cost and lower-limb positive work during level-ground and incline walking compared with walking without wearing the exoskeleton. This advancement bridges the gap between in-lab exoskeleton technology and real-world human ambulation, making exoskeleton control technology viable for a broad community.

Best linear and quadratic moments for spatial econometric models with an application to spatial interdependence patterns of employment growth in US counties

SummaryWe provide a novel analytic procedure to construct best linear and quadratic moments of the generalized method of moments estimation for a large class of cross‐sectional network and spatial econometric models. These moments generate an estimator that is asymptotically more efficient than the quasi‐maximum likelihood estimator when the disturbances follow a non‐normal and unknown distribution. We apply this procedure to a high‐order spatial autoregressive model with spatial errors, where the disturbances are heteroskedastic. Two normality tests of disturbances are developed. We apply the model to employment data in US counties, which demonstrates spatial interdependence patterns of regional employment growth.

Statistical Inference in Circular Structural Model and Fitting Circles to Noisy Data

It is well known that commonly used algorithms for circle fitting perform poorly when sampling distribution of the points is not symmetric with respect to the circle center, for example, when the points are sampled from a circle arc. To overcome this difficulty we introduce and study a parametric circular structural model. In this model the points on the circumference are assumed to be sampled according to the von Mises distribution with unknown concentration and mean direction parameters. We develop maximum likelihood and method of moments estimators of the circle center and radius, and study their statistical properties. In particular, we show that the proposed maximum likelihood estimator is asymptotically normal and efficient. We also develop a test of uniformity for the sampling distribution along the circle. Based on the derived theoretical results we propose a numerically stable circle fitting algorithm, investigate its accuracy in a simulation study, and illustrate its behavior in a real data example. Supplementary materials for this article are available online.

Exploring the Convergence of Macroeconomic Variables and Market Capitalization in an Emerging Economy: An Empirical Analysis Using Arbitrage Pricing Theory

This study delves into the interaction between macroeconomic variables and market capitalization in Emerging Capital Markets, i.e., Bangladesh. We examine the effects of the invest-to-deposit ratio (IDR) by financial institutions, market interest rate (MKTINT), and Consumer Price Index (CPI) on market capitalization (MAKTCAP). This study utilizes a dataset spanning 83 months, encompassing the period from 2015 to 2021. The macroeconomic influence is examined empirically by using the Arellano–Bover/Blundell–Bond generalized method of moments (GMM) estimator, supplemented by the ordinary least squares (OLS) and weighted least squares regression (WLS). Also, use Breusch-Pagan and Cook-Weisberg tests to measure heteroskedasticity in the model. The findings show a strong correlation between IDR and MAKTCAP, pointing to IDR as the most important yet overlooked factor influencing capital market returns. However, MKTINT and CPI results address a negative relationship with MAKTCAP. Higher IDR indicates that the capital market receives more cash flow to enhance market capitalization. At the same time, lowering MKTINT and CPI enables the capital market to get more money flow.