A Covariance Regression Model
By Peter D. Hoff, Xiaoyue Niu. 2011
122 citations recorded by the source corpus.
Classical regression analysis relates the expectation of a response vari- able to a linear combination of explanatory variables. In this article, we propose a covariance regression model that parameterizes the covariance matrix of a mul- tivariate response vector as a parsimonious quadratic function of explanatory vari- ables. The approach is analogous to the mean regression model, and is similar to a factor analysis model in which the factor loadings depend on the explanatory vari- ables. Using a random-effects representation, parameter estimation for the model is straightforward using either an EM-algorithm or an MCMC approximation via Gibbs sampling. The proposed methodology provides a simple but flexible repre- sentation of heteroscedasticity across the levels of an explanatory variable, improves estimation of the mean function and gives better calibrated prediction regions when compared to a homoscedastic model.
Subfields: Artificial Intelligence, Finance, Statistics and Probability
View Paper: https://www.semanticscholar.org/paper/d642f5a24fcf92ed435484450cc7a027a311b73c
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Mohsen Pourahmadi
- Wagner Hugo Bonat
- Bent Jørgensen
- Bruce A. Craig
- Anindya Bhadra
- P. R. Hansen
- I. Archakov
In Finance
In Statistics and Probability
- Mohsen Pourahmadi
- Clifford Lam
- Tommaso Cai
- Wei Biao Wu
- Bala Rajaratnam
- Kshitij Khare
- Harrison H. Zhou
- Joshua C. C. Chan
Citations in the corpus
- Covariance Estimation: The GLM and Regularization Perspectives (2011, background): “For example, Huang et al. (2006), Banerjee, El Ghaoui and d’Aspremont (2008), Friedman, Hastie and Tibshirani (2008), Rothman et al. (2008) and Warton (2008), respectively, impose penalty on the Cholesky factor, all the entries, off-diagonal entries and the diagonal entries of the precision matrix.”
- Covariance Regression Analysis (2017, unknown): “The second type directly links to the covariates under special model structures (e.g., see Prentice 1988; Demidenko 2004; Hoff and Niu 2012).”