How To Find A Valid Covariance Matrix
In some applications (e.g. building data models from only partially observed data) one wants to find the “nearest” covariance matrix to a given symmetric matrix (e.g. of observed covariances). In 2002, Higham formalized the notion of nearness using a weighted Frobenius norm and provided a method for computing the nearest covariance matrix.
Read more about this topic: Covariance Matrix
Famous quotes containing the words find, valid and/or matrix:
““If our minds could get hold of one abstract truth, they would be immortal so far as that truth is concerned. My trouble is to find out how we can get hold of the truth at all.””
—Henry Brooks Adams (1838–1918)
“Written poetry is worth reading once, and then should be destroyed. Let the dead poets make way for others. Then we might even come to see that it is our veneration for what has already been created, however beautiful and valid it may be, that petrifies us.”
—Antonin Artaud (1896–1948)
“In all cultures, the family imprints its members with selfhood. Human experience of identity has two elements; a sense of belonging and a sense of being separate. The laboratory in which these ingredients are mixed and dispensed is the family, the matrix of identity.”
—Salvador Minuchin (20th century)