检索结果-内蒙古大学图书馆

African Journal of Applied Statistics 2023年第1期10卷 1435-1450页

作者： Bruno Enagnon Lokonon Paulette Marthe Béhanzin Guedezoumè Scholastique M. Merveille Essetcheou Romain Glèlè Kakaï

generalized linear models ( GLM s) provide a unified approach that encompasses most of the linear models. However, they could not been properly used in agricultural sciences. In this review, we (1) examined the correctness related to the application of GLM s to agricultural sciences; and (2) provided a guideline for their suitable use. From a total of 832 articles identified, 155 articles were retained for the review. The results revealed that the use of GLM s in agricultural sciences has increased over time. However, in most cases (57.42%), GLM s outputs were not well interpreted. Moreover, useful expected findings from the use of GLM s were not reported in several papers. Indeed, 27.95% of the selected articles did not assess the goodness of the models and 21.30% did not report the significance of the coefficients. Our work has revealed the gap between the introduction of a statistical method and effective use in real world situations. A practical guideline with regard to GLM s has been provided. Les modèles linéaires généralisés ( GLM s) constituent une approche unifiée qui englobe la plupart des modèles linéaires. Cependant, ils pourraient être mal utilisés dans les sciences agricoles. Dans cette revue, nous avons (1) examiné l'exactitude de l'application des GLM s aux sciences agricoles et (2) fourni une ligne directrice pour leur utilisation appropriée. Sur un total de 832 articles identifiés, 155 ont été retenus pour l'examen. Les résultats ont révélé que l'utilisation des GLM s dans les sciences agricoles a augmenté au fil du temps. Toutefois, dans la plupart des cas (57,42 %), les résultats des GLM s n'ont pas été bien interprétés. En outre, les résultats utiles attendus de l'utilisation des GLM s n'ont pas été rapportés dans plusieurs articles. Par exemple, 27,95 % des articles sélectionnés n'ont pas évalué la qualité des modèles. Notre travail a révélé l'écart entre l'introduction d'une méthode statistique et son utilisation effective dans des situations

关键词： 62J12 92B05 92B15 agronomy critical review effective use generalized linear models non-normal error structures

来源：评论

学校读者我要写书评

暂无评论

The robust desparsified lasso and the focused information criterion for high-dimensional generalized linear models

引用

Series Statistics 2023年第1期57卷 1-25页

作者： S. C. Pandhare T. V. Ramanathan Department of Statistics Savitribai Phule Pune University Pune Maharashtra India

The classical lasso estimation for sparse, high-dimensional regression models is typically biased and lacks the oracle properties. The desparsified versions of the lasso have been proposed in the literature that attempt to overcome these drawbacks. In this paper, we propose the outliers-robust version of the desparsified lasso for high dimensional generalized linear models. The robustness, consistency and high dimensional asymptotics are investigated rigorously in a general framework ofM-estimation under potential model misspecification. The desparsification mechanism is subsequently utilized to construct the focused information criterion (FIC) thereby facilitating robust, focused model selection in high dimensions. The applications are demonstrated with the Poisson regression under robust quasilikelihood loss function. The empirical performance of the proposed methods is examined via simulations and a real data example.

关键词： Desparsified lasso focused information criterion generalized linear models high-dimensional asymptotics nodewise regression robust estimation

来源：评论

学校读者我要写书评

暂无评论

Predicting inappropriate source use from scores of language use, source comprehension, and organizational features: A study using generalized linear models

引用

ASSESSING WRITING 2025年 64卷

作者： Lee, Kwangmin Liao, Ray J. T. Hsiao, I. -Chun Vera Park, Junhee Ye, Yafei Western Michigan Univ Dept Special Educ & Literacy Studies Kalamazoo MI USA Natl Taiwan Ocean Univ Inst Appl English Keelung Taiwan Univ Iowa Dept Teaching & Learning Lowa IA USA Univ Iowa Dept Psychol & Quantitat Fdn Iowa City IA USA Zhengzhou Univ Sch Int Studies Zhengzhou Peoples R China

This paper examines the extent to which inappropriate source use - verbatim source use and patchwriting - can be predicted by scores of other textual features that are commonly evaluated in second/foreign language (L2) integrated writing assessment. 246 advanced-level English as a Foreign Language (EFL) test-takers enrolled in a Chinese higher education institution provided integrated essays that required both summary and argumentation. All the collected essays were rated by two experienced raters and checked for interrater reliability by way of generalizability theory. Then, a series of generalized linear models was compared to identify the best-fitting model that explained the relationship between the independent variables and inappropriate source use. Results indicated that the zero-inflated beta-binomial provided the best fit to the data, with approximately 43.67% of the data estimated to be an extra zero. Next, parameter estimates of this model included (1) non-significant effects of language use and source comprehension on inappropriate source use and (2) a significantly negative effect of organizational features on the dependent variable. This suggests that focusing on organizational features, operationalized herein as organization, coherence, development of ideas, and authorial voice, can help L2 test-takers reduce reliance on inappropriate source use. Implications for research and practice are discussed.

关键词： Verbatim source use Integrated assessment Second language writing generalized linear models

来源：评论

学校读者我要写书评

暂无评论

generalized linear models to assess environmental drivers of paralytic shellfish toxin blooms (Southeast Tasmania, Australia)

引用

CONTINENTAL SHELF RESEARCH 2021年 223卷 104439-104439页

作者： Xie, Hongrui Fischer, Andrew M. Strutton, Peter G. Univ Tasmania Inst Marine & Antarctic Studies Hobart Tas 7001 Australia Univ Tasmania Inst Marine & Antarctic Studies Launceston Tas 7250 Australia Univ Tasmania Australian Res Council Ctr Excellence Climate Ext Hobart Tas 7001 Australia

Paralytic shellfish toxin (PST) events are often difficult to predict because synergistic processes, involving multiple temporal and spatial scales, are responsible for their development. In this study, we consider a suite of factors, from local to regional processes, driving paralytic shellfish toxin (PST) events associated with Alexandrium catenella on the east coast of Tasmania (Australia). PST concentrations in shellfish were compared against environmental variables, including daily wind speed and direction, satellite chlorophyll-a, sea surface temperature and river discharge. Relationships were tested using generalized linear models (GLMs). GLM coefficients confirm that high PST events were significantly favored by cooler temperatures and higher chlorophyll-a concentrations. PST events were seasonal, occurring between August and November and showed a significant relationship with sustained coastal upwelling and wind transition events after one to three months of predominantly upwelling winds. Satellite imagery shows evidence of cross shelf transport moderating PST events. Overall, the results suggest that a combination of factors (currents, upwelling conditions and changes in sea surface temperature) work together to initiate and moderate PST events. The use of GLMs provides a useful tool for understanding synergistic relationships. These techniques are of value for developing reliable predictive models to forecast and manage the effects of harmful and costly PST events.

关键词： Paralytic shellfish toxins Sea surface temperature Chlorophyll-a Upwelling relaxation generalized linear models

来源：评论

学校读者我要写书评

暂无评论

Functional Analysis of generalized linear models Under Nonlinear Constraints with Artificial Intelligence and Machine Learning Applications to the Sciences

Functional Analysis of Generalized Linear Models Under Nonli...

引用

作者： Chowdhury, K. P. University of California Irvine

学位级别：Ph.D., Doctor of Philosophy

This thesis presents multiple fundamental mathematical contributions to generalized linear models (GLMs) ubiquitous to the sciences. The methodologies considered are shown to overcome biased estimates for parameters of interest in the sciences through new mathematical results and their applications in both nonparametric and parametric settings. The results are shown to be uniformly better in comparison to existing widely used methods in the sciences. In extensive simulation studies the methodologies outperform existing Artificial Intelligence (AI) and Machine Learning (ML) methods in the sciences for all around better Model fits, Inference and Prediction (MIP) results without losing interpretability of the parameter estimates. This is because the mathematical construction and their accompanying mathematical foundations ensure that the estimation procedure strongly converges to the parameters of interest. In the first application, I present a parametric version of the methodology (© Elsevier and Journal of Informetrics) titled “Functional analysis of generalized linear models under non-linear constraints with applications to identifying highly-cited papers.” In the second application, I extend this methodology in an entirely nonparametric setting which gives equivalent results to the parametric formulation under various circumstances, but may outperform it as well in others, especially if the underlying Data Generating Process (DGP) is asymmetric. Furthermore, I show that the categorical data models on which the methodologies are applied can be extended to any GLM, continuous or otherwise, while maintaining model interpretability and convergence results. In addition, I present a new prediction performance diagnostic statistic, called Adjusted ROC Statistic (ARS), which allows us to compare whether the prediction performance of various models fitted are statistically different. The nonparametric methodology is then further extended to give a new formulation of the bin

关键词： Adjusted ROC Statistic generalized linear models Estimation algorithm

来源：评论

学校读者我要写书评

暂无评论

Inverse probability of treatment weighting with generalized linear outcome models for doubly robust estimation

引用

STATISTICS IN MEDICINE 2024年第3期43卷 534-547页

作者： Gabriel, Erin E. Sachs, Michael C. Martinussen, Torben Waernbaum, Ingeborg Goetghebeur, Els Vansteelandt, Stijn Sjolander, Arvid Univ Copenhagen Dept Publ Hlth Sect Biostat Copenhagen Denmark Uppsala Univ Dept Stat Uppsala Sweden Univ Ghent Dept Appl Math Comp Sci & Stat Ghent Belgium Karolinska Inst Dept Med Epidemiol & Biostat Stockholm Sweden

There are now many options for doubly robust estimation;however, there is a concerning trend in the applied literature to believe that the combination of a propensity score and an adjusted outcome model automatically results in a doubly robust estimator and/or to misuse more complex established doubly robust estimators. A simple alternative, canonical link generalized linear models (GLM) fit via inverse probability of treatment (propensity score) weighted maximum likelihood estimation followed by standardization (the g$$ g $$-formula) for the average causal effect, is a doubly robust estimation method. Our aim is for the reader not just to be able to use this method, which we refer to as IPTW GLM, for doubly robust estimation, but to fully understand why it has the doubly robust property. For this reason, we define clearly, and in multiple ways, all concepts needed to understand the method and why it is doubly robust. In addition, we want to make very clear that the mere combination of propensity score weighting and an adjusted outcome model does not generally result in a doubly robust estimator. Finally, we hope to dispel the misconception that one can adjust for residual confounding remaining after propensity score weighting by adjusting in the outcome model for what remains 'unbalanced' even when using doubly robust estimators. We provide R code for our simulations and real open-source data examples that can be followed step-by-step to use and hopefully understand the IPTW GLM method. We also compare to a much better-known but still simple doubly robust estimator.

关键词： causal inference doubly robust generalized linear models

来源：评论

学校读者我要写书评

暂无评论

Complete subset averaging approach for high-dimensional generalized linear models

引用

ECONOMICS LETTERS 2023年 226卷

作者： Chen, Xingyi Li, Haiqi Zhang, Jing Hunan Univ Coll Finance & Stat Changsha 410006 Peoples R China

This study proposes a novel complete subset averaging (CSA) method for high-dimensional generalized linear models based on a penalized Kullback-Leibler (KL) loss. All models under consideration can be potentially misspecified, and the dimension of covariates is allowed to diverge to infinity. The uniform convergence rate and asymptotic normality of the proposed estimator are established. Moreover, it is asymptotically optimal in terms of achieving the lowest KL loss. To ease the computational burden, we randomly draw a fixed number of subsets from the complete subsets and show their asymptotic equivalence. The Monte Carlo simulation and empirical application demonstrate that the proposed CSA method outperforms popular model-averaging methods.(c) 2023 Published by Elsevier B.V.

关键词： Asymptotic optimality Complete subset averaging Kullback-Leibler loss generalized linear models

来源：评论

学校读者我要写书评

暂无评论

Dual Extrapolation for Sparse generalized linear models

引用

JOURNAL OF MACHINE LEARNING RESEARCH 2020年 21卷 1-33页

作者： Massias, Mathurin Vaiter, Samuel Gramfort, Alexandre Salmon, Joseph Univ Paris Saclay CEA INRIA Palaiseau France CNRS F-21078 Dijon France Inst Math Bourgogne F-21078 Dijon France Univ Montpellier CNRS IMAG F-34095 Montpellier France

generalized linear models (GLM) form a wide class of regression and classification models, where prediction is a function of a linear combination of the input variables. For statistical inference in high dimension, sparsity inducing regularizations have proven to be useful while offering statistical guarantees. However, solving the resulting optimization problems can be challenging: even for popular iterative algorithms such as coordinate descent, one needs to loop over a large number of variables. To mitigate this, techniques known as screening rules and working sets diminish the size of the optimization problem at hand, either by progressively removing variables, or by solving a growing sequence of smaller problems. For both techniques, significant variables are identified thanks to convex duality arguments. In this paper, we show that the dual iterates of a GLM exhibit a Vector AutoRegressive (VAR) behavior after sign identification, when the primal problem is solved with proximal gradient descent or cyclic coordinate descent. Exploiting this regularity, one can construct dual points that offer tighter certificates of optimality, enhancing the performance of screening rules and working set algorithms.

关键词： Convex optimization extrapolation screening rules working sets Lasso sparse logistic regression generalized linear models

来源：评论

学校读者我要写书评

暂无评论

Approximate and exact optimal designs for 2^k factorial experiments for generalized linear models via second order cone programming

引用

STATISTICAL PAPERS 2020年第6期61卷 2737-2767页

作者： Duarte, Belmiro P. M. Sagnol, Guillaume Inst Super Engn Coimbra Inst Politecn Coimbra Dept Chem & Biol Engn Rua Pedro Nunes P-3030199 Coimbra Portugal Univ Coimbra Dept Chem Engn CIEPQPF Coimbra Portugal Tech Univ Berlin Inst Math Berlin Germany

Model-based optimal designs of experiments (M-bODE) for nonlinear models are typically hard to compute. The literature on the computation of M-bODE for nonlinear models when the covariates are categorical variables, i.e. factorial experiments, is scarce. We propose second order cone programming (SOCP) and Mixed Integer Second Order Programming (MISOCP) formulations to find, respectively, approximate and exact A- and D-optimal designs for 2(k) factorial experiments for generalized linear models (GLMs). First, locally optimal (approximate and exact) designs for GLMs are addressed using the formulation of Sagnol (J Stat Plan Inference 141(5):1684-1708, 2011). Next, we consider the scenario where the parameters are uncertain, and new formulations are proposed to find Bayesian optimal designs using the A- and log detD-optimality criteria. A quasi Monte-Carlo sampling procedure based on the Hammersley sequence is used for computing the expectation in the parametric region of interest. We demonstrate the application of the algorithm with the logistic, probit and complementary log-log models and consider full and fractional factorial designs.

关键词： D-optimal designs 2(k) Factorial experiments Exact designs Second order cone programming generalized linear models Quasi-Monte Carlo sampling

来源：评论

学校读者我要写书评

暂无评论

Reflection on modern methods: generalized linear models for prognosis and intervention-theory, practice and implications for machine learning

引用

INTERNATIONAL JOURNAL OF EPIDEMIOLOGY 2020年第6期49卷 2074-2082页

作者： Arnold, Kellyn F. Davies, Vinny de Kamps, Marc Tennant, Peter W. G. Mbotwa, John Gilthorpe, Mark S. Univ Leeds Leeds Inst Data Analyt Level 11 Worsley BldgClarendon Way Leeds LS2 9NL W Yorkshire England Univ Leeds Sch Med Leeds W Yorkshire England Univ Glasgow Sch Comp Sci Glasgow Lanark Scotland Univ Leeds Sch Comp Leeds W Yorkshire England Alan Turing Inst London England

Prediction and causal explanation are fundamentally distinct tasks of data analysis. In health applications, this difference can be understood in terms of the difference between prognosis (prediction) and prevention/treatment (causal explanation). Nevertheless, these two concepts are often conflated in practice. We use the framework of generalized linear models (GLMs) to illustrate that predictive and causal queries require distinct processes for their application and subsequent interpretation of results. In particular, we identify five primary ways in which GLMs for prediction differ from GLMs for causal inference: (i) the covariates that should be considered for inclusion in (and possibly exclusion from) the model;(ii) how a suitable set of covariates to include in the model is determined;(iii) which covariates are ultimately selected and what functional form (i.e. parameterization) they take;(iv) how the model is evaluated;and (v) how the model is interpreted. We outline some of the potential consequences of failing to acknowledge and respect these differences, and additionally consider the implications for machine learning (ML) methods. We then conclude with three recommendations that we hope will help ensure that both prediction and causal modelling are used appropriately and to greatest effect in health research.

关键词： Prediction causal inference generalized linear models directed acyclic graphs machine learning artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：