Regression from dependent observations

DOI10.1145/3313276.3316362MaRDI QIDQ5212828zbMATH OpenOpenAlexFDO

Authors Constantinos Daskalakis, Nishanth Dikkala, Ioannis Panageas

Publication date 30 January 2020

Published in Proceedings of the 51st Annual ACM SIGACT Symposium on Theory of Computing (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1905.03353

zbMATH Keywords

estimationGaussian graphical modelsIsing modelslinear regressionlogistic regression

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Probabilistic graphical models (62H22) Linear regression; mixed models (62J05) Generalized linear models (logistic models) (62J12)

Abstract: The standard linear and logistic regression models assume that the response variables are independent, but share the same linear relationship to their corresponding vectors of covariates. The assumption that the response variables are independent is, however, too strong. In many applications, these responses are collected on nodes of a network, or some spatial or temporal domain, and are dependent. Examples abound in financial and meteorological applications, and dependencies naturally arise in social networks through peer effects. Regression with dependent responses has thus received a lot of attention in the Statistics and Economics literature, but there are no strong consistency results unless multiple independent samples of the vectors of dependent responses can be collected from these models. We present computationally and statistically efficient methods for linear and logistic regression models when the response variables are dependent on a network. Given one sample from a networked linear or logistic regression model and under mild assumptions, we prove strong consistency results for recovering the vector of coefficients and the strength of the dependencies, recovering the rates of standard regression under independent observations. We use projected gradient descent on the negative log-likelihood, or negative log-pseudolikelihood, and establish their strong convexity and consistency using concentration of measure for dependent random variables.

Recommendations

• Limit theorems for network dependent random variables
• Logistic Regression for Dependent Binary Observations
• Regression analysis of networked data
• Fixed‐Effect Regressions on Network Data
• Errors-in-variables models with dependent measurements

Cited in

(17)

• Dependence by reverse regular rule
• Network regression with predictive clustering trees
• Limit theorems for network dependent random variables
• Fluctuations of the magnetization in the \(p\)-spin Curie-Weiss model
• A linear regression with unobserved dependent variables
• Regression analysis of networked data
• Sub-critical exponential random graphs: concentration of measure and some applications
• From association to causation via regression
• Benign overfitting in time-series linear models with over-parameterization
• Causal effect estimation under network interference with mean-field methods
• Nonsense associations in Markov random fields with pairwise dependence
• One-shot learning for k-SAT
• Limit theorems and phase transitions in the tensor Curie-Weiss Potts model
• High dimensional logistic regression under network dependence
• Phase transitions of the maximum likelihood estimators in the p-spin Curie-Weiss model
• One-shot learning for \(k\)-SAT
• Regression with strongly correlated data

This page was built for publication: Regression from dependent observations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5212828)