scholarly journals Benefits of dimension reduction in penalized regression methods for high-dimensional grouped data: a case study in low sample size

2019 ◽  
Vol 35 (19) ◽  
pp. 3628-3634 ◽  
Author(s):  
Soufiane Ajana ◽  
Niyazi Acar ◽  
Lionel Bretillon ◽  
Boris P Hejblum ◽  
Hélène Jacqmin-Gadda ◽  
...  
Keyword(s):  
Dimension Reduction ◽  
Sample Size ◽  
High Dimension ◽  
Penalized Regression ◽  
Group Lasso ◽  
Supplementary Information ◽  
Grouped Data ◽  
Regression Methods ◽  
Additional Information

Abstract Motivation In some prediction analyses, predictors have a natural grouping structure and selecting predictors accounting for this additional information could be more effective for predicting the outcome accurately. Moreover, in a high dimension low sample size framework, obtaining a good predictive model becomes very challenging. The objective of this work was to investigate the benefits of dimension reduction in penalized regression methods, in terms of prediction performance and variable selection consistency, in high dimension low sample size data. Using two real datasets, we compared the performances of lasso, elastic net, group lasso, sparse group lasso, sparse partial least squares (PLS), group PLS and sparse group PLS. Results Considering dimension reduction in penalized regression methods improved the prediction accuracy. The sparse group PLS reached the lowest prediction error while consistently selecting a few predictors from a single group. Availability and implementation R codes for the prediction methods are freely available at https://github.com/SoufianeAjana/Blisar. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Scoping Review of Thermal Comfort Research in Colombia

Buildings ◽  
2021 ◽  
Vol 11 (6) ◽  
pp. 232
Author(s):  
Juan Manuel Medina ◽  
Carolina M. Rodriguez ◽  
Maria Camila Coronado ◽  
Lina Maria Garcia
Keyword(s):  
Thermal Comfort ◽  
Scoping Review ◽  
Architectural Design ◽  
Urban Climate ◽  
Academic Community ◽  
Additional Information ◽  
The Future ◽  
Art Research ◽  
Occupant Satisfaction

The analysis of thermal comfort in buildings, energy consumption, and occupant satisfaction is crucial to influencing the architectural design methodologies of the future. However, research in these fields in developing countries is sectorised. Most times, the standards to study and assess thermal comfort such as ASHRAE Standard 55, EN 15251, and ISO 7730 are insufficient and not appropriate for the geographical areas of application. This article presents a scoping review of published work in Colombia, as a representative case study, to highlight the state-of-the-art, research trends, gaps, and potential areas for further development. It examines the amount, origin, extent, and content of research and peer-reviewed documentation over the last decades. The findings allow new insights regarding the preferred models and the evaluation tools that have been used to date and that are recommended to use in the future. It also includes additional information regarding the most and least studied regions, cities, and climates in the country. This work could be of interest for the academic community and policymakers in the areas related to indoor and urban climate management and energy efficiency.

scholarly journals Mapping Complex Land Use Histories and Urban Renewal Using Ground Penetrating Radar: A Case Study from Fort Stanwix

2021 ◽  
Vol 13 (13) ◽  
pp. 2478
Author(s):  
Tyler Stumpf ◽  
Daniel P. Bigman ◽  
Dominic J. Day
Keyword(s):  
Urban Expansion ◽  
Educational Programming ◽  
Additional Information ◽  
National Monument ◽  
Early Colonial ◽  
Historic Structures ◽  
Economic Revitalization ◽  
Management Plans ◽  
Ground Penetrating

Fort Stanwix National Monument, located in Rome, NY, is a historic park with a complex use history dating back to the early Colonial period and through the urban expansion and recent economic revitalization of the City of Rome. The goal of this study was to conduct a GPR investigation over an area approximately 1 acre in size to identify buried historic features (particularly buildings) so park management can preserve these resources and develop appropriate educational programming and management plans. The GPR recorded reflection events consistent with our expectations of historic structures. Differences in size, shape, orientation, and depth suggest that these responses likely date to different time periods in the site’s history. The GPR recorded other reflection anomalies that are difficult to interpret without any additional information, which suggests that pairing high-density geophysical data with limited excavations is critical to elaborate a complex site’s intricate history.

scholarly journals Parallelized calculation of permutation tests

2020 ◽  
Author(s):  
Markus Ekvall ◽  
Michael Höhle ◽  
Lukas Käll
Keyword(s):  
Dynamic Programming ◽  
Sample Size ◽  
Permutation Test ◽  
Statistical Tests ◽  
Permutation Tests ◽  
Supplementary Information ◽  
Attractive Alternative ◽  
Test Statistic ◽  
Sample Distribution ◽  
Running Time

Abstract Motivation Permutation tests offer a straightforward framework to assess the significance of differences in sample statistics. A significant advantage of permutation tests are the relatively few assumptions about the distribution of the test statistic are needed, as they rely on the assumption of exchangeability of the group labels. They have great value, as they allow a sensitivity analysis to determine the extent to which the assumed broad sample distribution of the test statistic applies. However, in this situation, permutation tests are rarely applied because the running time of naïve implementations is too slow and grows exponentially with the sample size. Nevertheless, continued development in the 1980s introduced dynamic programming algorithms that compute exact permutation tests in polynomial time. Albeit this significant running time reduction, the exact test has not yet become one of the predominant statistical tests for medium sample size. Here, we propose a computational parallelization of one such dynamic programming-based permutation test, the Green algorithm, which makes the permutation test more attractive. Results Parallelization of the Green algorithm was found possible by non-trivial rearrangement of the structure of the algorithm. A speed-up—by orders of magnitude—is achievable by executing the parallelized algorithm on a GPU. We demonstrate that the execution time essentially becomes a non-issue for sample sizes, even as high as hundreds of samples. This improvement makes our method an attractive alternative to, e.g. the widely used asymptotic Mann-Whitney U-test. Availabilityand implementation In Python 3 code from the GitHub repository https://github.com/statisticalbiotechnology/parallelPermutationTest under an Apache 2.0 license. Supplementary information Supplementary data are available at Bioinformatics online.

Robust centroid based classification with minimum error rates for high dimension, low sample size data

2009 ◽  
Vol 139 (8) ◽  
pp. 2571-2580 ◽  
Author(s):  
Jiancheng Jiang ◽  
J.S. Marron ◽  
Xuejun Jiang
Keyword(s):  
Sample Size ◽  
High Dimension ◽  
Error Rates ◽  
Minimum Error ◽  
Size Data

Sample size and informetric model goodness-of-fit outcomes: a search engine log case study

2006 ◽  
Vol 32 (3) ◽  
pp. 212-222 ◽  
Author(s):  
Isola Ajiferuke ◽  
Dietmar Wolfram ◽  
Felix Famoye
Keyword(s):  
Sample Size ◽  
Search Engine ◽  
Goodness Of Fit

Penalized Regression Methods in the Source Analysis of Face Recognition

2008 ◽  
pp. 621-624
Author(s):  
Mayrim Vega-Hernández ◽  
Eduardo Martínez-Montes ◽  
Jhoanna Pérez-Hidalgo-Gato ◽  
José M. Sánchez-Bornot ◽  
Pedro Valdés-Sosa
Keyword(s):  
Face Recognition ◽  
Penalized Regression ◽  
Source Analysis ◽  
Regression Methods
Sign in / Sign up

Export Citation Format

Share Document