The development of methods for the meta-analysis of diagnostic test accuracy (DTA) studies is still an active area of research. While methods for the standard case where each study reports a single pair of sensitivity and specificity are nearly routinely applied nowadays, methods to meta-analyze receiver operating characteristic (ROC) curves are not widely used. This situation is more complex, as each primary DTA study may report on several pairs of sensitivity and specificity, each corresponding to a different threshold. In a case study published earlier, we applied a number of methods for meta-analyzing DTA studies with multiple thresholds to a real-world data example (Zapf et al., Biometrical Journal. 2021; 63(4): 699–711). To date, no simulation study exists that systematically compares different approaches with respect to their performance in various scenarios when the truth is known. In this article, we aim to fill this gap and present the results of a simulation study that compares three frequentist approaches for the meta-analysis of ROC curves. We performed a systematic simulation study, motivated by an example from medical research. In the simulations, all three approaches worked partially well. The approach by Hoyer and colleagues was slightly superior in most scenarios and is recommended in practice.
{"title":"Meta-Analysis of Diagnostic Accuracy Studies With Multiple Thresholds: Comparison of Approaches in a Simulation Study","authors":"Antonia Zapf, Cornelia Frömke, Juliane Hardt, Gerta Rücker, Dina Voeltz, Annika Hoyer","doi":"10.1002/bimj.202300101","DOIUrl":"https://doi.org/10.1002/bimj.202300101","url":null,"abstract":"<p>The development of methods for the meta-analysis of diagnostic test accuracy (DTA) studies is still an active area of research. While methods for the standard case where each study reports a single pair of sensitivity and specificity are nearly routinely applied nowadays, methods to meta-analyze receiver operating characteristic (ROC) curves are not widely used. This situation is more complex, as each primary DTA study may report on several pairs of sensitivity and specificity, each corresponding to a different threshold. In a case study published earlier, we applied a number of methods for meta-analyzing DTA studies with multiple thresholds to a real-world data example (Zapf et al., <i>Biometrical Journal</i>. 2021; 63(4): 699–711). To date, no simulation study exists that systematically compares different approaches with respect to their performance in various scenarios when the truth is known. In this article, we aim to fill this gap and present the results of a simulation study that compares three frequentist approaches for the meta-analysis of ROC curves. We performed a systematic simulation study, motivated by an example from medical research. In the simulations, all three approaches worked partially well. The approach by Hoyer and colleagues was slightly superior in most scenarios and is recommended in practice.</p>","PeriodicalId":55360,"journal":{"name":"Biometrical Journal","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/bimj.202300101","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142324593","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The fit of a regression model to new data is often worse due to overfitting. Analysts use variable selection techniques to develop parsimonious regression models, which may introduce bias into regression estimates. Shrinkage methods have been proposed to mitigate overfitting and reduce bias in estimates. Post-estimation shrinkage is an alternative to penalized methods. This study evaluates effectiveness of post-estimation shrinkage in improving prediction performance of full and selected models. Through a simulation study, results were compared with ordinary least squares (OLS) and ridge in full models, and best subset selection (BSS) and lasso in selected models. We focused on prediction errors and the number of selected variables. Additionally, we proposed a modified version of the parameter-wise shrinkage (PWS) approach named non-negative PWS (NPWS) to address weaknesses of PWS. Results showed that no method was superior in all scenarios. In full models, NPWS outperformed global shrinkage, whereas PWS was inferior to OLS. In low correlation with moderate-to-high signal-to-noise ratio (SNR), NPWS outperformed ridge, but ridge performed best in small sample sizes, high correlation, and low SNR. In selected models, all post-estimation shrinkage performed similarly, with global shrinkage slightly inferior. Lasso outperformed BSS and post-estimation shrinkage in small sample sizes, low SNR, and high correlation but was inferior when the opposite was true. Our study suggests that, with sufficient information, NPWS is more effective than global shrinkage in improving prediction accuracy of models. However, in high correlation, small sample sizes, and low SNR, penalized methods generally outperform post-estimation shrinkage methods.
{"title":"Post-Estimation Shrinkage in Full and Selected Linear Regression Models in Low-Dimensional Data Revisited","authors":"Edwin Kipruto, Willi Sauerbrei","doi":"10.1002/bimj.202300368","DOIUrl":"https://doi.org/10.1002/bimj.202300368","url":null,"abstract":"<p>The fit of a regression model to new data is often worse due to overfitting. Analysts use variable selection techniques to develop parsimonious regression models, which may introduce bias into regression estimates. Shrinkage methods have been proposed to mitigate overfitting and reduce bias in estimates. Post-estimation shrinkage is an alternative to penalized methods. This study evaluates effectiveness of post-estimation shrinkage in improving prediction performance of full and selected models. Through a simulation study, results were compared with ordinary least squares (OLS) and ridge in full models, and best subset selection (BSS) and lasso in selected models. We focused on prediction errors and the number of selected variables. Additionally, we proposed a modified version of the parameter-wise shrinkage (PWS) approach named non-negative PWS (NPWS) to address weaknesses of PWS. Results showed that no method was superior in all scenarios. In full models, NPWS outperformed global shrinkage, whereas PWS was inferior to OLS. In low correlation with moderate-to-high signal-to-noise ratio (SNR), NPWS outperformed ridge, but ridge performed best in small sample sizes, high correlation, and low SNR. In selected models, all post-estimation shrinkage performed similarly, with global shrinkage slightly inferior. Lasso outperformed BSS and post-estimation shrinkage in small sample sizes, low SNR, and high correlation but was inferior when the opposite was true. Our study suggests that, with sufficient information, NPWS is more effective than global shrinkage in improving prediction accuracy of models. However, in high correlation, small sample sizes, and low SNR, penalized methods generally outperform post-estimation shrinkage methods.</p>","PeriodicalId":55360,"journal":{"name":"Biometrical Journal","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/bimj.202300368","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142324583","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jan Gertheiss, David Rügamer, Bernard X. W. Liew, Sonja Greven
Functional data analysis (FDA) is a statistical framework that allows for the analysis of curves, images, or functions on higher dimensional domains. The goals of FDA, such as descriptive analyses, classification, and regression, are generally the same as for statistical analyses of scalar-valued or multivariate data, but FDA brings additional challenges due to the high- and infinite dimensionality of observations and parameters, respectively. This paper provides an introduction to FDA, including a description of the most common statistical analysis techniques, their respective software implementations, and some recent developments in the field. The paper covers fundamental concepts such as descriptives and outliers, smoothing, amplitude and phase variation, and functional principal component analysis. It also discusses functional regression, statistical inference with functional data, functional classification and clustering, and machine learning approaches for functional data analysis. The methods discussed in this paper are widely applicable in fields such as medicine, biophysics, neuroscience, and chemistry and are increasingly relevant due to the widespread use of technologies that allow for the collection of functional data. Sparse functional data methods are also relevant for longitudinal data analysis. All presented methods are demonstrated using available software in R by analyzing a dataset on human motion and motor control. To facilitate the understanding of the methods, their implementation, and hands-on application, the code for these practical examples is made available through a code and data supplement and on GitHub.
函数数据分析(FDA)是一种统计框架,可用于分析高维域上的曲线、图像或函数。函数数据分析的目标,如描述性分析、分类和回归,与标量值或多变量数据统计分析的目标大致相同,但由于观测值和参数分别具有高维和无限维,函数数据分析带来了额外的挑战。本文介绍了 FDA,包括最常见的统计分析技术、各自的软件实现以及该领域的一些最新进展。本文涵盖了一些基本概念,如描述值和离群值、平滑、振幅和相位变化以及函数主成分分析。论文还讨论了功能回归、功能数据统计推断、功能分类和聚类,以及用于功能数据分析的机器学习方法。本文讨论的方法可广泛应用于医学、生物物理学、神经科学和化学等领域,而且由于可收集功能数据的技术的广泛应用,这些方法的相关性日益增强。稀疏功能数据方法也适用于纵向数据分析。通过分析人类运动和运动控制的数据集,使用现有的 R 软件演示了所有介绍的方法。为了便于理解这些方法、实现这些方法以及进行实际应用,我们通过代码和数据补充以及 GitHub 提供了这些实际示例的代码。
{"title":"Functional Data Analysis: An Introduction and Recent Developments","authors":"Jan Gertheiss, David Rügamer, Bernard X. W. Liew, Sonja Greven","doi":"10.1002/bimj.202300363","DOIUrl":"https://doi.org/10.1002/bimj.202300363","url":null,"abstract":"<p>Functional data analysis (FDA) is a statistical framework that allows for the analysis of curves, images, or functions on higher dimensional domains. The goals of FDA, such as descriptive analyses, classification, and regression, are generally the same as for statistical analyses of scalar-valued or multivariate data, but FDA brings additional challenges due to the high- and infinite dimensionality of observations and parameters, respectively. This paper provides an introduction to FDA, including a description of the most common statistical analysis techniques, their respective software implementations, and some recent developments in the field. The paper covers fundamental concepts such as descriptives and outliers, smoothing, amplitude and phase variation, and functional principal component analysis. It also discusses functional regression, statistical inference with functional data, functional classification and clustering, and machine learning approaches for functional data analysis. The methods discussed in this paper are widely applicable in fields such as medicine, biophysics, neuroscience, and chemistry and are increasingly relevant due to the widespread use of technologies that allow for the collection of functional data. Sparse functional data methods are also relevant for longitudinal data analysis. All presented methods are demonstrated using available software in R by analyzing a dataset on human motion and motor control. To facilitate the understanding of the methods, their implementation, and hands-on application, the code for these practical examples is made available through a code and data supplement and on GitHub.</p>","PeriodicalId":55360,"journal":{"name":"Biometrical Journal","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/bimj.202300363","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142324584","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}