From 36539eac7a492e9c52823e6aa4ecb013a9cdb79d Mon Sep 17 00:00:00 2001 From: Roger Herikstad Date: Fri, 10 Mar 2023 10:51:18 +0800 Subject: [PATCH] Fixes formula for the covariance of an FA model The second **W** factor, rather than the first, should be transposed. --- docs/source/fa.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/source/fa.rst b/docs/source/fa.rst index c068fc5..2fa1c9d 100644 --- a/docs/source/fa.rst +++ b/docs/source/fa.rst @@ -52,7 +52,7 @@ or use it to reconstruct (approximately) the observations from latent variables, \tilde{\mathbf{x}} = \mathbf{\Sigma} \mathbf{W} (\mathbf{W}^T \mathbf{W})^{-1} \mathbf{z} + \boldsymbol{\mu} -Here, :math:`\mathbf{W}` is the factor loadings or weight matrix, :math:`\mathbf{\Sigma} = \mathbf{\Psi} + \mathbf{W}^T \mathbf{W}` is the covariance matrix. +Here, :math:`\mathbf{W}` is the factor loadings or weight matrix, :math:`\mathbf{\Sigma} = \mathbf{\Psi} + \mathbf{W} \mathbf{W}^T` is the covariance matrix. The package provides methods to do so: