Sparse recovery under matrix uncertainty
Rosenbaum, Mathieu ; Tsybakov, Alexandre B.
Ann. Statist., Tome 38 (2010) no. 1, p. 2620-2651 / Harvested from Project Euclid
We consider the model y = Xθ + ξ, Z = X + Ξ, ¶ where the random vector y ∈ ℝn and the random n × p matrix Z are observed, the n × p matrix X is unknown, Ξ is an n × p random noise matrix, ξ ∈ ℝn is a noise independent of Ξ, and θ is a vector of unknown parameters to be estimated. The matrix uncertainty is in the fact that X is observed with additive error. For dimensions p that can be much larger than the sample size n, we consider the estimation of sparse vectors θ. Under matrix uncertainty, the Lasso and Dantzig selector turn out to be extremely unstable in recovering the sparsity pattern (i.e., of the set of nonzero components of θ), even if the noise level is very small. We suggest new estimators called matrix uncertainty selectors (or, shortly, the MU-selectors) which are close to θ in different norms and in the prediction risk if the restricted eigenvalue assumption on X is satisfied. We also show that under somewhat stronger assumptions, these estimators recover correctly the sparsity pattern.
Publié le : 2010-10-15
Classification:  Sparsity,  MU-selector,  matrix uncertainty,  errors-in-variables model,  measurement error,  sign consistency,  oracle inequalities,  restricted eigenvalue assumption,  missing data,  portfolio selection,  portfolio replication,  62J05,  62F12
@article{1278861455,
     author = {Rosenbaum, Mathieu and Tsybakov, Alexandre B.},
     title = {Sparse recovery under matrix uncertainty},
     journal = {Ann. Statist.},
     volume = {38},
     number = {1},
     year = {2010},
     pages = { 2620-2651},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1278861455}
}
Rosenbaum, Mathieu; Tsybakov, Alexandre B. Sparse recovery under matrix uncertainty. Ann. Statist., Tome 38 (2010) no. 1, pp.  2620-2651. http://gdmltest.u-ga.fr/item/1278861455/