Journal of Applied Statistics 2020
5 papers
In 2020, the Journal of Applied Statistics published research focused on advancing regression techniques, experimental design, and time series analysis. Key themes included the development of robust estimation methods and the integration of sparsity in machine learning models. For instance, studies explored regularized robust estimation within binary regression models and introduced structured sparse support vector machines designed for data with ordered features. These contributions reflect a trend toward enhancing the reliability and interpretability of statistical models in complex data environments.
Methodological innovations also extended to experimental design and compositional data analysis. Notable work included the application of maximum expected entropy to transform Latin hypercube designs and the use of partial least squares regression for compositional response variables and covariates. Additionally, researchers investigated parametric bootstrap techniques for testing correlations between time series. Prolific contributors during this period included authors such as Qingguo Tang, Rohana J. Karunamuni, Thomas J. Fisher, and Karel Hron. Collectively, the 2020 edition emphasized computational efficiency and the refinement of diagnostic tools for diverse statistical applications.