结果(博弈论)
选择(遗传算法)
疾病
营养流行病学
秩(图论)
统计
先验与后验
计量经济学
统计假设检验
回归分析
回归
流行病学
计算机科学
医学
数学
机器学习
病理
数理经济学
哲学
组合数学
认识论
作者
Cornelia Weikert,Matthias B. Schulze
标识
DOI:10.1097/mco.0000000000000308
摘要
The purpose of review is to present methodological issues as well as most relevant recent developments on the application of a statistical method to derive dietary patterns: reduced rank regression (RRR). RRR can be used efficiently in nutritional epidemiology to identify dietary patterns associated with selected response variables that have known relations with a disease outcome of interest. This has the advantage of building on a priori knowledge of biological relations, by including plausible intermediates between diet and the outcome of interest.This statistical method has been applied first in nutritional epidemiology about 1 decade ago. Since then, more than 60 publications were published applying the RRR. This method is considerably dependent on an adequate selection of response variables. These response sets were most often a combination of nutrients or of selected endogenous biomarkers. But also variables of intermediate clinical phenotype or contaminants were selected. However, applying this method, several methodological issues, such as, for example, selection of responses, simplification, and validation of the derived pattern should be taken into account.RRR is a modern statistical method to derive dietary patterns that can be used to test specific hypothesis on pathways from diet to development of a disease.
科研通智能强力驱动
Strongly Powered by AbleSci AI