scispace - formally typeset
Journal ArticleDOI

A new approach to null correlations of proportions

John Aitchison
- 01 Apr 1981 - 
- Vol. 13, Iss: 2, pp 175-189
Reads0
Chats0
TLDR
In this article, the dependence structure of proportions in terms of such concepts has been investigated, and an alternative method of modeling and distribution theory and tests for dependence analysis has been proposed.
Abstract
Much work on the statistical analysis of compositional data has concentrated on the difficulty of interpreting correlations between proportions with an assortment of tests for nullcorrelations, for independence except for the constraint, F-independence of bounded variables, neutrality in the mean and in the median. This paper questions the appropriateness of characterizing the dependence structure of proportions in terms of such concepts, suggests an alternative method of modeling, develops necessary distribution theory and tests, and illustrates the methodology in applications.

read more

Citations
More filters
Journal ArticleDOI

Inferring Correlation Networks from Genomic Survey Data

TL;DR: SparCC as mentioned in this paper is a new approach that is capable of estimating correlation values from compositional data, which is used to infer a rich ecological network connecting hundreds of interacting species across 18 sites on the human body.
Journal ArticleDOI

The Statistical Analysis of Compositional Data

TL;DR: In this article, the authors present an approach to perform compositional analysis of geochemical compositions of rocks using logratio linear models and a combination of matrix covariance analysis and linear linear models.
Journal ArticleDOI

Microbial Co-occurrence Relationships in the Human Microbiome

TL;DR: This article applied an ensemble method based on multiple similarity measures in combination with generalized boosted linear models (GBLMs) to taxonomic marker (16S rRNA gene) profiles of this cohort, resulting in a global network of 3,005 significant co-occurrence and co-exclusion relationships between 197 clades occurring throughout the human microbiome.

Microbial Co-occurrence Relationships in the Human Microbiome

TL;DR: An ensemble method based on multiple similarity measures in combination with generalized boosted linear models (GBLMs) to taxonomic marker (16S rRNA gene) profiles of this cohort resulted in a global network of 3,005 significant co-occurrence and co-exclusion relationships between 197 clades occurring throughout the human microbiome.
Journal ArticleDOI

Sparse and Compositionally Robust Inference of Microbial Ecological Networks

TL;DR: SParse InversE Covariance Estimation for Ecological Association Inference is presented, a statistical method for the inference of microbial ecological networks from amplicon sequencing datasets that outperforms state-of-the-art methods to recover edges and network properties on synthetic data under a variety of scenarios.
References
More filters
Journal ArticleDOI

The Lognormal Distribution.

TL;DR: Lloyds Bank has its main root in a substantial private bank founded in Birmingham nearly two centuries ago; one hundred years ago this Bank still had only the one office in Birmingham, with a related private banking house in Lombard Street, and by amalgamation it has absorbed scores of other eighteenth and nineteenth century banks, both private and joint stock, and at least two of the former reach back into Restoration London, perhaps Cromwellian London.
Book

The lognormal distribution

TL;DR: Lloyds Bank has its main root in a substantial private bank founded in Birmingham nearly two centuries ago; one hundred years ago this Bank still had only the one office in Birmingham, with a related private banking house in Lombard Street, and by amalgamation it has absorbed scores of other eighteenth and nineteenth century banks, both private and joint stock, and at least two of the former reach back into Restoration London, perhaps Cromwellian London.
Journal ArticleDOI

On the Distribution of the Likelihood Ratio

TL;DR: In this paper, the asymptotic distribution of the likelihood ratio λ is examined when the value of the parameter is a boundary point of both the set of points corresponding to the hypothesis and the set corresponding to an alternative.
Related Papers (5)