Article
Social Sciences, Interdisciplinary
Monica Pratesi, Luciana Quattrociocchi, Gaia Bertarelli, Alessandro Gemignani, Caterina Giusti
Summary: The concept of educational poverty, defined as lack of educational opportunities and rights, has attracted the interest of researchers due to its complexities and consequences. Italy's National Statistical Institute has started measuring educational poverty through the composite educational poverty index (EPI), using small area estimation models to understand its spatial distribution at a local level and help policymakers allocate resources effectively.
SOCIAL INDICATORS RESEARCH
(2021)
Article
Computer Science, Interdisciplinary Applications
Jan Pablo Burgard, Joscha Krause, Simon Schmaus
Summary: Spatial dynamic microsimulations allow for multivariate analysis of complex systems with accurate simulation outcomes. However, the reliance on survey data for transition probabilities can be challenging due to lack of regional detail and coverage, potentially leading to simulation failure.
COMPUTATIONAL STATISTICS & DATA ANALYSIS
(2021)
Article
Multidisciplinary Sciences
Cuong Viet Nguyen, Khuong Duc Nguyen, Tuyen Quang Tran
Summary: Our study reveals significant spatial heterogeneity in electric power consumption between districts and provinces in Vietnam using a small area estimation method. Households in mountains and highlands consume less electricity compared to those in delta and coastal areas. Additionally, a U-shaped relationship between electricity consumption inequality and economic levels in Vietnam is found.
Article
Multidisciplinary Sciences
Cuong Viet Nguyen, Khuong Duc Nguyen, Tuyen Quang Tran
Summary: Our study applies a small area estimation method to estimate the average and inequality of per capita kWh consumption in Vietnam. The findings reveal a significant spatial heterogeneity in electric power consumption between districts and provinces, with households in mountainous and highland areas consuming considerably less electricity than those in delta and coastal areas. Importantly, there is a U-shaped relationship between electricity consumption inequality and economic levels in Vietnam, with higher inequality in poor districts and provinces and lower inequality in middle-income areas.
Article
Computer Science, Interdisciplinary Applications
Christian H. Weiss, Murat Caner Testik, Annika Homburg
Summary: This study is a first step towards analyzing the effects of parameter estimation on the monitoring of autocorrelated count processes, focusing on factors like dispersion and model structures. The results suggest that the dispersion of the count process is a key parameter in chart design, and for low-dispersion scenarios, the impact of parameter estimation on design parameters may vanish.
COMPUTERS & INDUSTRIAL ENGINEERING
(2021)
Article
Computer Science, Interdisciplinary Applications
Paul Corral Rodas, Isabel Molina, Minh Nguyen
Summary: This paper presents a methodological update to the World Bank's toolkit for small area estimation, with revised methods yielding considerably less biased and more efficient estimators compared to the clustered bootstrap approach.
JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION
(2021)
Article
Energy & Fuels
Marc Mari-Dell'Olmo, Laura Oliveras, Carlos Vergara-Hernandez, Lucia Artazcoz, Carme Borrell, Merce Gotsens, Laia Palencia, Maria Jose Lopez, Miguel A. Martinez-Beneito
Summary: Energy poverty is a growing issue in urban areas of southern Europe, including Barcelona. This study analyzed geographical inequalities in energy poverty in Barcelona by estimating small-area indicators and an index. The results revealed significant disparities in the distribution of energy poverty, with certain neighborhoods being more affected. These findings emphasize the need for targeted interventions and policy-making to alleviate energy poverty.
Article
Engineering, Electrical & Electronic
Jiahao Liu, Cheng Wang, Tianshu Bi, Guoyi Xu
Summary: This paper proposes an approach to online estimate power system inertia, which is essential for frequency stability analysis and control. The method can estimate the aggregated inertia at the point of interconnection (POI) and calculate the power system total inertia as well as monitor the inertia spatial distribution. Based on the spatial correlation of the power system frequency, the general correlation between the rotor speed and network bus frequency is derived. The proposed method considers field implementation scenario with ambient measurement data and relaxed requirement for mechanical power measurement.
IEEE TRANSACTIONS ON POWER SYSTEMS
(2023)
Article
Green & Sustainable Science & Technology
Haojian Dou, Llibang Ma, Shichun Liu, Fang Fang
Summary: The accurate identification of poverty types in rural areas, using Gansu Province in China as an example, reveals that the multidimensional poverty index in the region presents a pattern of high at both ends and low in the middle spatially. Factors such as the proportion of highway entrances and exits and water resource management are identified as main contributors to poverty.
ENVIRONMENT DEVELOPMENT AND SUSTAINABILITY
(2022)
Article
Engineering, Electrical & Electronic
Qiqi Zhu, Jiale Chen, Linlin Wang, Qingfeng Guan
Summary: This article proposes a scene unmixing framework based on nonnegative matrix factorization for urban mixed scenes (UnUMS) to estimate the mixing ratio of urban scenes. To confirm the effectiveness, datasets of different scales were utilized for experiments, and the results demonstrate that UnUMS can accurately estimate the proportions of different scenes.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Computer Science, Interdisciplinary Applications
Prateek Bansal, Rico Krueger, Daniel J. Graham
Summary: Spatial count data models are used to predict phenomena frequencies in geographically distinct entities. Variational Bayes method offers faster estimation and better performance in both simulation studies and empirical applications.
COMPUTATIONAL STATISTICS & DATA ANALYSIS
(2021)
Article
Social Sciences, Mathematical Methods
Carolina Franco, William R. Bell
Summary: This study demonstrates how estimates from the American Community Survey (ACS) can be used to improve estimates from smaller household surveys in the US. By utilizing bivariate area-level models, the authors exploit the strong relationships between population characteristics estimated by the smaller surveys and ACS estimates. The study shows impressive reductions in variance for state estimates of health insurance coverage and county estimates of poverty of school-aged children. The models achieve large variance reductions even without using regression covariates drawn from auxiliary data sources.
JOURNAL OF SURVEY STATISTICS AND METHODOLOGY
(2022)
Article
Computer Science, Information Systems
Qianqian Zhou, Nan Chen, Siwei Lin
Summary: This paper proposes a new poverty measurement model based on a neural network to capture the spatial correlation of poverty and improve the accuracy of poverty identification. The experiments show that the model outperforms other baseline models in measuring the poverty index and the measured index is highly consistent with the actual situation. This model is of great significance for poverty tracking and resource allocation in developing countries.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2022)
Article
Social Sciences, Mathematical Methods
Naomi Diz-Rosales, Maria Jose Lombardia, Domingo Morales
Summary: This article presents a statistical methodology based on a random regression coefficient Poisson model for predicting counts and proportions in small areas, and introduces bootstrap estimators for mean squared errors. The Laplace approximation algorithm is used to calculate maximum likelihood estimators of model parameters and mode predictors of random effects. Simulation experiments demonstrate the behavior of the fitting algorithm, predictors, and MSE estimators with and without bias correction. The methodology is applied to data from the Spanish Living Conditions Survey to estimate the proportions of women and men below the poverty line by province.
JOURNAL OF SURVEY STATISTICS AND METHODOLOGY
(2023)
Article
Multidisciplinary Sciences
Elizaveta Semenova, Yidan Xu, Adam Howes, Theo Rashid, Samir Bhatt, Swapnil Mishra, Seth Flaxman
Summary: Gaussian processes are widely used in small-area spatial statistical modelling, but their computational challenges limit their scalability and practical usefulness. This paper proposes a novel deep generative modelling approach called PriorVAE, which approximates Gaussian process priors through prior sampling and variational autoencoder fitting, improving the efficiency of spatial inference.
JOURNAL OF THE ROYAL SOCIETY INTERFACE
(2022)
Article
Statistics & Probability
Miguel Boubeta, Maria Jose Lombardia, Domingo Morales
Article
Statistics & Probability
Miguel Boubeta, Maria Jose Lombardia, Domingo Morales
Article
Social Sciences, Mathematical Methods
Maria Jose Lombardia, Esther Lopez-Vizcaino, Cristina Rueda
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES A-STATISTICS IN SOCIETY
(2017)
Article
Statistics & Probability
Maria Dolores Esteban, Maria Jose Lombardia, Esther Lopez-Vizcaino, Domingo Morales, Agustin Perez
Article
Statistics & Probability
Maria Jose Lombardia, Esther Lopez-Vizcaino, Cristina Rueda
Summary: The paper introduces a small area estimation approach that borrows strength across domains and time, using data across time to select different models for each domain and obtaining estimators with good performance through aggregated mixed AIC statistic.
Article
Statistics & Probability
Maria Dolores Esteban, Maria Jose Lombardia, Esther Lopez-Vizcaino, Domingo Morales, Agustin Perez
Summary: This paper introduces a unit-level bivariate linear mixed model and analyzes the behavior of fitting algorithm, predictors, and mean squared error estimators through simulation experiments and the application of real data.
JOURNAL OF APPLIED STATISTICS
(2022)
Article
Statistics & Probability
Katarzyna Reluga, Maria-Jose Lombardia, Stefan Sperlich
Summary: This paper focuses on the development of a framework for simultaneous inference in generalized linear mixed models (GLMM) and applies it to widely used examples of mixed models. Through extensive simulations and a case study, the effectiveness and advantages of the simultaneous inference tools are demonstrated.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2023)
Article
Social Sciences, Mathematical Methods
Maria Jose Lombardia, Esther Lopez-Vizcaino, Cristina Rueda
Summary: This paper presents an original approach to estimate the gender pay gap, incorporating model selection and bias correction, and has been validated in a small area context. The proposed method shows good performance and scalability, making it a promising tool for analyzing wage differentials.
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES A-STATISTICS IN SOCIETY
(2022)
Article
Statistics & Probability
Katarzyna Reluga, Maria-Jose Lombardia, Stefan Sperlich
Summary: Linear mixed models have gained significant attention in various fields of applied statistics in recent decades, particularly when dealing with clustered, hierarchical, or longitudinal data. However, there is a scarcity of statistical tools for valid simultaneous inference for mixed parameters. This paper discusses methods for simultaneous inference in linear mixed models, including the development of simultaneous prediction intervals and multiple testing procedures for mixed parameters.
INTERNATIONAL STATISTICAL REVIEW
(2022)
Article
Statistics & Probability
Maria Dolores Esteban, Maria Jose Lombardia, Esther Lopez-Vizcaino, Domingo Morales, Agustin Perez
Summary: This paper introduces a statistical methodology based on nested error regression model for predicting small area bivariate parameters. The corresponding mean squared errors are estimated using parametric bootstrap and the behavior of the introduced methodology is empirically studied through several simulation experiments. An application to real data from the Spanish household budget survey provides estimators of ratios of food household expenditures by provinces.
SCANDINAVIAN JOURNAL OF STATISTICS
(2022)
Article
Statistics & Probability
Maria Dolores Esteban, Maria Jose Lombardia, Esther Lopez-Vizcaino, Domingo Morales, Agustin Perez
Summary: This paper investigates the small area estimation of population averages of unit-level compositional data. It proposes a new methodology to transform compositions into vectors and assumes a multivariate nested error regression model. Empirical best predictors of domain indicators are derived and their mean squared errors are estimated. Simulation experiments are conducted to investigate the behavior of the introduced predictors. An application to real data from the Spanish household budget survey is given.
Article
Forestry
Miguel Boubeta, Maria Jose Lombardia, Manuel Marey-Perez, Domingo Morales
INTERNATIONAL JOURNAL OF WILDLAND FIRE
(2019)
Article
Statistics & Probability
Maria Jose Lombardia, Esther Lopez-Vizcaino, Cristina Rueda
STATISTICS AND APPLICATIONS
(2018)
Article
Computer Science, Interdisciplinary Applications
Miguel Boubeta, Maria Jose Lombardia, Domingo Morales
COMPUTATIONAL STATISTICS & DATA ANALYSIS
(2017)