Pub Date : 2025-03-28DOI: 10.1016/j.spasta.2025.100895
Frederic Schoenberg
Current methods for evaluating earthquake forecasts, such as the -test, -test, or log-likelihood score, typically do not disproportionately reward a model for more accurately forecasting the largest events, or disproportionately punish a model for less accurately forecasting the largest events. However, since the largest earthquakes are by far the most destructive and therefore of most interest to practitioners, in many circumstances, a weighted likelihood score may be more useful. Here, we propose various weighted measures, weighting each earthquake by some function of its magnitude, such as potency-weighted log-likelihood, and consider their properties. The proposed methods are applied to a catalog of earthquakes in the Western United States.
{"title":"Magnitude-weighted goodness-of-fit scores for earthquake forecasting","authors":"Frederic Schoenberg","doi":"10.1016/j.spasta.2025.100895","DOIUrl":"10.1016/j.spasta.2025.100895","url":null,"abstract":"<div><div>Current methods for evaluating earthquake forecasts, such as the <span><math><mi>N</mi></math></span>-test, <span><math><mi>L</mi></math></span>-test, or log-likelihood score, typically do not disproportionately reward a model for more accurately forecasting the largest events, or disproportionately punish a model for less accurately forecasting the largest events. However, since the largest earthquakes are by far the most destructive and therefore of most interest to practitioners, in many circumstances, a weighted likelihood score may be more useful. Here, we propose various weighted measures, weighting each earthquake by some function of its magnitude, such as potency-weighted log-likelihood, and consider their properties. The proposed methods are applied to a catalog of earthquakes in the Western United States.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"67 ","pages":"Article 100895"},"PeriodicalIF":2.1,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143746591","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-27DOI: 10.1016/j.spasta.2025.100899
Murat Okatan
Background
Spatio-temporal receptive fields (STRFs) of visual neurons are often estimated using spike-triggered averaging (STA) with binary pseudo-random stimulus sequences. An exact analytical test—called the STA-BPRS test—has been developed to determine the statistical significance of each pixel in the STRF estimate. However, computing this test can take minutes to days, or even longer, for certain neurons.
New method
Here, the STA-BPRS test is accelerated by approximating the null distribution of STRF pixel estimates with a Normal distribution. This methodological refinement significantly reduces computation time, making large-scale data analysis feasible.
Results
The approximate test is systematically validated on real mouse retinal ganglion cell data and synthetic spike train data, demonstrating that it yields identical significance thresholds to the exact test. For neurons where, exact computation would be prohibitively long (e.g., hundreds of years), the approximate test completes in seconds or minutes.
Comparison with existing methods
Few approaches address pixel-by-pixel significance in STA-based STRF estimates. While subspace methods like spike-triggered covariance exist for STRF estimation, they typically do not provide direct voxel-wise or pixel-wise p-values. The proposed method specifically accelerates an exact distribution-based test.
Conclusions and impact
The proposed Normal approximation drastically reduces computation time, enabling high-throughput analysis of STRF mapping from spike data. This advancement may foster broader adoption of precise statistical tests of STRFs in large-scale, high-density electrophysiological recordings. Moreover, fast detection of significant STRF features could facilitate closed-loop experiments where stimuli dynamically adapt to changing STRF structures.
{"title":"Fast computation of the statistical significance test for spatio-temporal receptive field estimates obtained using spike-triggered averaging of binary pseudo-random sequences","authors":"Murat Okatan","doi":"10.1016/j.spasta.2025.100899","DOIUrl":"10.1016/j.spasta.2025.100899","url":null,"abstract":"<div><h3>Background</h3><div>Spatio-temporal receptive fields (STRFs) of visual neurons are often estimated using spike-triggered averaging (STA) with binary pseudo-random stimulus sequences. An exact analytical test—called the STA-BPRS test—has been developed to determine the statistical significance of each pixel in the STRF estimate. However, computing this test can take minutes to days, or even longer, for certain neurons.</div></div><div><h3>New method</h3><div>Here, the STA-BPRS test is accelerated by approximating the null distribution of STRF pixel estimates with a Normal distribution. This methodological refinement significantly reduces computation time, making large-scale data analysis feasible.</div></div><div><h3>Results</h3><div>The approximate test is systematically validated on real mouse retinal ganglion cell data and synthetic spike train data, demonstrating that it yields identical significance thresholds to the exact test. For neurons where, exact computation would be prohibitively long (e.g., hundreds of years), the approximate test completes in seconds or minutes.</div></div><div><h3>Comparison with existing methods</h3><div>Few approaches address pixel-by-pixel significance in STA-based STRF estimates. While subspace methods like spike-triggered covariance exist for STRF estimation, they typically do not provide direct voxel-wise or pixel-wise p-values. The proposed method specifically accelerates an exact distribution-based test.</div></div><div><h3>Conclusions and impact</h3><div>The proposed Normal approximation drastically reduces computation time, enabling high-throughput analysis of STRF mapping from spike data. This advancement may foster broader adoption of precise statistical tests of STRFs in large-scale, high-density electrophysiological recordings. Moreover, fast detection of significant STRF features could facilitate closed-loop experiments where stimuli dynamically adapt to changing STRF structures.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"67 ","pages":"Article 100899"},"PeriodicalIF":2.1,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143767964","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-26DOI: 10.1016/j.spasta.2025.100894
Kinspride Duah, Yan Sun, Brennan Bean
Environmental data are often imprecise due to various limitations and uncertainties in the measuring process. As a result, they often consist of a combination of both precise and imprecise information, referred to as hard and soft data, respectively. Often in practice, soft data are characterized as intervals as a simple form to properly preserve the underlying imprecision. Bayesian maximum entropy (BME) is a generalized spatial interpolation method that processes both hard and soft data simultaneously to effectively account for both spatial uncertainty and measurement imprecision. This paper presents a rigorous evaluation to compare the performances of BME and kriging through both simulation and a case study of reliability-targeted design ground snow load (RTDSL) prediction in Utah. The dataset contains a mixture of hard and soft-interval observations, and kriging uses the soft-interval data by extracting the midpoints in addition to the hard data. The cross-validated results show that BME outperforms kriging on multiple error metrics. Specifically for hard data locations where precise observations are known, BME yields a mean error (ME) of 0.0334, a mean absolute error (MAE) of 0.2309, and a root mean squared error (RMSE) of 0.2833, whereas kriging produces a ME of 0.1960, MAE of 0.2793, and RMSE of 0.3698. These results highlight the superior prediction accuracy of BME, particularly in the presence of soft data and/or non-Gaussian hard data.
{"title":"Integrating multi-source geospatial information using Bayesian maximum entropy: A case study on design ground snow load prediction","authors":"Kinspride Duah, Yan Sun, Brennan Bean","doi":"10.1016/j.spasta.2025.100894","DOIUrl":"10.1016/j.spasta.2025.100894","url":null,"abstract":"<div><div>Environmental data are often imprecise due to various limitations and uncertainties in the measuring process. As a result, they often consist of a combination of both precise and imprecise information, referred to as hard and soft data, respectively. Often in practice, soft data are characterized as intervals as a simple form to properly preserve the underlying imprecision. Bayesian maximum entropy (BME) is a generalized spatial interpolation method that processes both hard and soft data simultaneously to effectively account for both spatial uncertainty and measurement imprecision. This paper presents a rigorous evaluation to compare the performances of BME and kriging through both simulation and a case study of reliability-targeted design ground snow load (RTDSL) prediction in Utah. The dataset contains a mixture of hard and soft-interval observations, and kriging uses the soft-interval data by extracting the midpoints in addition to the hard data. The cross-validated results show that BME outperforms kriging on multiple error metrics. Specifically for hard data locations where precise observations are known, BME yields a mean error (ME) of 0.0334, a mean absolute error (MAE) of 0.2309, and a root mean squared error (RMSE) of 0.2833, whereas kriging produces a ME of 0.1960, MAE of 0.2793, and RMSE of 0.3698. These results highlight the superior prediction accuracy of BME, particularly in the presence of soft data and/or non-Gaussian hard data.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"67 ","pages":"Article 100894"},"PeriodicalIF":2.1,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143739998","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-13DOI: 10.1016/j.spasta.2025.100891
S. Ward, E.A.K. Cohen, N.M. Adams
The fundamental functional summary statistics used for studying spatial point patterns are developed for marked homogeneous and inhomogeneous point processes on the surface of a sphere. These are extended to point processes on the surface of three dimensional convex shapes given the bijective mapping from the shape to the sphere is known. These functional summary statistics are used to test for independence between the marginals of multi-type spatial point processes with methods for sampling the null distribution developed and discussed. This is illustrated on both simulated data and the RNGC galaxy point pattern, revealing attractive dependencies between different galaxy types.
{"title":"Functional summary statistics and testing for independence in multi-type point processes on the surface of three dimensional convex shapes","authors":"S. Ward, E.A.K. Cohen, N.M. Adams","doi":"10.1016/j.spasta.2025.100891","DOIUrl":"10.1016/j.spasta.2025.100891","url":null,"abstract":"<div><div>The fundamental functional summary statistics used for studying spatial point patterns are developed for marked homogeneous and inhomogeneous point processes on the surface of a sphere. These are extended to point processes on the surface of three dimensional convex shapes given the bijective mapping from the shape to the sphere is known. These functional summary statistics are used to test for independence between the marginals of multi-type spatial point processes with methods for sampling the null distribution developed and discussed. This is illustrated on both simulated data and the RNGC galaxy point pattern, revealing attractive dependencies between different galaxy types.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"67 ","pages":"Article 100891"},"PeriodicalIF":2.1,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143679461","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-12DOI: 10.1016/j.spasta.2025.100892
Miao Long, Zhimeng Sun
This paper presents a Bayesian adaptive Lasso estimation approach for partially linear hierarchical spatial autoregressive models. Despite advancements in spatial modeling, two key gaps remain: the lack of non-linear components in hierarchical spatial autoregressive models to capture complex spatial relationships, and the insufficient application of dimensionality reduction techniques to address high-dimensionality and overfitting. This paper addresses these issues by combining partially linear models with spatial autoregressive structures and incorporating dimensionality reduction techniques to enhance model efficiency and mitigate overfitting. The hierarchical structure facilitates multi-level modeling, accommodating complex data relationships. The Bayesian adaptive Lasso technique ensures effective variable selection and regularization, improving model interpretability and performance. Simulations and real data applications demonstrate the proposed method’s excellent performance. This work offers valuable insights for researchers and practitioners in dealing with spatially correlated data in various fields.
{"title":"Bayesian adaptive Lasso estimation for partially linear hierarchical spatial autoregressive model","authors":"Miao Long, Zhimeng Sun","doi":"10.1016/j.spasta.2025.100892","DOIUrl":"10.1016/j.spasta.2025.100892","url":null,"abstract":"<div><div>This paper presents a Bayesian adaptive Lasso estimation approach for partially linear hierarchical spatial autoregressive models. Despite advancements in spatial modeling, two key gaps remain: the lack of non-linear components in hierarchical spatial autoregressive models to capture complex spatial relationships, and the insufficient application of dimensionality reduction techniques to address high-dimensionality and overfitting. This paper addresses these issues by combining partially linear models with spatial autoregressive structures and incorporating dimensionality reduction techniques to enhance model efficiency and mitigate overfitting. The hierarchical structure facilitates multi-level modeling, accommodating complex data relationships. The Bayesian adaptive Lasso technique ensures effective variable selection and regularization, improving model interpretability and performance. Simulations and real data applications demonstrate the proposed method’s excellent performance. This work offers valuable insights for researchers and practitioners in dealing with spatially correlated data in various fields.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"67 ","pages":"Article 100892"},"PeriodicalIF":2.1,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143627809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-27DOI: 10.1016/j.spasta.2025.100890
Guilherme Ludwig , Yuan Wang , Tingjin Chu , Haonan Wang , Jun Zhu
Statistical analysis has been successfully applied to crime data for identification of crime hot spots and prediction of future crimes. In this paper, our main objective is to identify key factors for gun violence in Rio de Janeiro and study the relationship between these key factors and the number of reported events. We use a Bayesian hierarchical stochastic Poisson regression model for spatial counts, which enables us to address the over-dispersed count data and to handle the spatial correlation. Moreover, we propose a variable selection method for key factor identification based on the spike-and-slab prior distribution for the regression coefficients. A new Gibbs sampler is developed for sampling from the posterior distributions with the help of augmentation of Pólya-Gamma auxiliary variables. Simulation studies are used to demonstrate the performance of our proposed approach. Our analysis of the gun violence data in Rio de Janeiro reveals the relationship between violence events and socio-demographic covariates as well as an interpretable spatial random effect that accounts for unmeasured covariate information.
{"title":"Bayesian analysis and variable selection for spatial count data with an application to Rio de Janeiro gun violence","authors":"Guilherme Ludwig , Yuan Wang , Tingjin Chu , Haonan Wang , Jun Zhu","doi":"10.1016/j.spasta.2025.100890","DOIUrl":"10.1016/j.spasta.2025.100890","url":null,"abstract":"<div><div>Statistical analysis has been successfully applied to crime data for identification of crime hot spots and prediction of future crimes. In this paper, our main objective is to identify key factors for gun violence in Rio de Janeiro and study the relationship between these key factors and the number of reported events. We use a Bayesian hierarchical stochastic Poisson regression model for spatial counts, which enables us to address the over-dispersed count data and to handle the spatial correlation. Moreover, we propose a variable selection method for key factor identification based on the spike-and-slab prior distribution for the regression coefficients. A new Gibbs sampler is developed for sampling from the posterior distributions with the help of augmentation of Pólya-Gamma auxiliary variables. Simulation studies are used to demonstrate the performance of our proposed approach. Our analysis of the gun violence data in Rio de Janeiro reveals the relationship between violence events and socio-demographic covariates as well as an interpretable spatial random effect that accounts for unmeasured covariate information.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"67 ","pages":"Article 100890"},"PeriodicalIF":2.1,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143519049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-24DOI: 10.1016/j.spasta.2025.100885
Claudio Rubino , Chiara Di Maria , Antonino Abbruzzo , Gioacchino Bono , Germana Garofalo , Giacomo Milisenda , Giada Adelfio
In many applied fields, it may be of interest to evaluate mediational mechanisms occurring in spatial domains. The approaches proposed so far in the literature to address this issue deal with areal data and often consider linear models. In this paper, we propose an approach to assess mediation in the presence of geostatistical data by combining the integrated nested Laplace approximation (INLA) with a derivative-based approach for mediation analysis, which allows one to estimate indirect effects also in the case of nonlinear models. We investigate the effect of ignoring spatial processes in the mediator and the outcome models through a simulation study, focusing also on the case of correlated processes. To show the usefulness of our approach, we also provided an ecological application.
{"title":"Derivative-based spatial mediation with INLA-SPDE","authors":"Claudio Rubino , Chiara Di Maria , Antonino Abbruzzo , Gioacchino Bono , Germana Garofalo , Giacomo Milisenda , Giada Adelfio","doi":"10.1016/j.spasta.2025.100885","DOIUrl":"10.1016/j.spasta.2025.100885","url":null,"abstract":"<div><div>In many applied fields, it may be of interest to evaluate mediational mechanisms occurring in spatial domains. The approaches proposed so far in the literature to address this issue deal with areal data and often consider linear models. In this paper, we propose an approach to assess mediation in the presence of geostatistical data by combining the integrated nested Laplace approximation (INLA) with a derivative-based approach for mediation analysis, which allows one to estimate indirect effects also in the case of nonlinear models. We investigate the effect of ignoring spatial processes in the mediator and the outcome models through a simulation study, focusing also on the case of correlated processes. To show the usefulness of our approach, we also provided an ecological application.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"66 ","pages":"Article 100885"},"PeriodicalIF":2.1,"publicationDate":"2025-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143488347","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Factor analysis has been extensively used to reveal the dependence structures among multivariate variables, offering valuable insight in various fields. However, it cannot incorporate the spatial heterogeneity that is typically present in spatial data. To address this issue, we introduce an effective method specifically designed to discover the potential dependence structures in multivariate spatial data. Our approach assumes that spatial locations can be approximately divided into a finite number of clusters, with locations within the same cluster sharing similar dependence structures. By leveraging an iterative algorithm that combines spatial clustering with factor analysis, we simultaneously detect spatial clusters and estimate a unique factor model for each cluster. The proposed method is evaluated through comprehensive simulation studies, demonstrating its flexibility. In addition, we apply the proposed method to a dataset of railway station attributes in the Tokyo metropolitan area, highlighting its practical applicability and effectiveness in uncovering complex spatial dependencies.
{"title":"Clustered factor analysis for multivariate spatial data","authors":"Yanxiu Jin , Tomoya Wakayama , Renhe Jiang , Shonosuke Sugasawa","doi":"10.1016/j.spasta.2025.100889","DOIUrl":"10.1016/j.spasta.2025.100889","url":null,"abstract":"<div><div>Factor analysis has been extensively used to reveal the dependence structures among multivariate variables, offering valuable insight in various fields. However, it cannot incorporate the spatial heterogeneity that is typically present in spatial data. To address this issue, we introduce an effective method specifically designed to discover the potential dependence structures in multivariate spatial data. Our approach assumes that spatial locations can be approximately divided into a finite number of clusters, with locations within the same cluster sharing similar dependence structures. By leveraging an iterative algorithm that combines spatial clustering with factor analysis, we simultaneously detect spatial clusters and estimate a unique factor model for each cluster. The proposed method is evaluated through comprehensive simulation studies, demonstrating its flexibility. In addition, we apply the proposed method to a dataset of railway station attributes in the Tokyo metropolitan area, highlighting its practical applicability and effectiveness in uncovering complex spatial dependencies.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"66 ","pages":"Article 100889"},"PeriodicalIF":2.1,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143508562","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-22DOI: 10.1016/j.spasta.2025.100888
Zaineb Smida , Thibault Laurent , Lionel Cucala
A scan method for functional data indexed in space has been developed. The scan statistic is derived from the Hotelling test statistic for functional data, extending the univariate and multivariate Gaussian spatial scan statistics. This method consistently outperforms existing techniques in detecting and locating spatial clusters, as demonstrated through simulations. It has been applied to two types of real data: economic data in order to identify spatial clusters of abnormal unemployment rates in Spain and climatic data in order to detect unusual climate change patterns in Great Britain, Nigeria, Pakistan, and Venezuela.
{"title":"A Hotelling spatial scan statistic for functional data: Application to economic and climate data","authors":"Zaineb Smida , Thibault Laurent , Lionel Cucala","doi":"10.1016/j.spasta.2025.100888","DOIUrl":"10.1016/j.spasta.2025.100888","url":null,"abstract":"<div><div>A scan method for functional data indexed in space has been developed. The scan statistic is derived from the Hotelling test statistic for functional data, extending the univariate and multivariate Gaussian spatial scan statistics. This method consistently outperforms existing techniques in detecting and locating spatial clusters, as demonstrated through simulations. It has been applied to two types of real data: economic data in order to identify spatial clusters of abnormal unemployment rates in Spain and climatic data in order to detect unusual climate change patterns in Great Britain, Nigeria, Pakistan, and Venezuela.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"66 ","pages":"Article 100888"},"PeriodicalIF":2.1,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143508312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-18DOI: 10.1016/j.spasta.2025.100887
Lingling Tian , Chuanhua Wei , Mixia Wu
This paper investigates a partially linear spatial autoregressive panel data model that incorporates fixed effects, constant and time-varying regression coefficients, and a time-varying spatial lag coefficient. A two-stage least squares estimation method based on profile local linear dummy variables (2SLS-PLLDV) is proposed to estimate both constant and time-varying coefficients without the need for first differencing. The asymptotic properties of the estimator are derived under certain conditions. Furthermore, a residual-based goodness-of-fit test is constructed for the model, and a residual-based bootstrap method is used to obtain p-values. Simulation studies show the good performance of the proposed method in various scenarios. For illustration, the carbon emission data from Chinese provinces and the public capital productivity data from the United States are analyzed.
{"title":"Statistical inference of partially linear time-varying coefficients spatial autoregressive panel data model","authors":"Lingling Tian , Chuanhua Wei , Mixia Wu","doi":"10.1016/j.spasta.2025.100887","DOIUrl":"10.1016/j.spasta.2025.100887","url":null,"abstract":"<div><div>This paper investigates a partially linear spatial autoregressive panel data model that incorporates fixed effects, constant and time-varying regression coefficients, and a time-varying spatial lag coefficient. A two-stage least squares estimation method based on profile local linear dummy variables (2SLS-PLLDV) is proposed to estimate both constant and time-varying coefficients without the need for first differencing. The asymptotic properties of the estimator are derived under certain conditions. Furthermore, a residual-based goodness-of-fit test is constructed for the model, and a residual-based bootstrap method is used to obtain p-values. Simulation studies show the good performance of the proposed method in various scenarios. For illustration, the carbon emission data from Chinese provinces and the public capital productivity data from the United States are analyzed.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"66 ","pages":"Article 100887"},"PeriodicalIF":2.1,"publicationDate":"2025-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143445043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号