Covariance estimation using modified PCA, based on BurStFin
Recap of the factor model:
R n x 1 returns; a n x 1 intercept; B n x m loading; G n x m 'factor-mimicking portfolios'; f m x 1 factor score; S n x n covariance of R;
R = a + B.f + e (1)
E[f.f'] = I(m) orthonormal factors; E[e.e'] = psi specific risk (diagonal)
S = B.B' + psi (2)
f = R.G (3)
The function fms2 estimates B,G; returns a list which is informally of class ce.
The remaining functions are accessors.
Package: | aace |
Type: | Package |
Version: | 1.0 |
Date: | 2013-04-16 |
License: \GPL (>=2) | |
Giles Heywood
Maintainer: gilesheywood at gmail.com
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.