$\newcommand{\var}{\text{Var}}$$\newcommand{\cov}{\text{Cov}}$$\newcommand{\E}{\text{E}}$$\newcommand{\one}{\mathbf 1}$$\newcommand{\0}{\mathbf 0}$$\newcommand{\a}{\alpha}$$\newcommand{\l}{\lambda}$$\newcommand{\s}{\sigma}$$\newcommand{\e}{\varepsilon}$$\newcommand{\t}{\theta}$In this post I’m going to consider a linear model of the form $y = \mu\one + Z\a…
$\newcommand{\a}{\alpha}$$\newcommand{\L}{\mathcal L}$$\newcommand{\one}{\mathbf 1}$$\newcommand{\x}{\mathbf x}$$\newcommand{\0}{\mathbf 0}$Let $G = (V,E)$ be an undirected and unweighted graph on $n$ vertices. I’ll have $A\in\{0,1\}^{n\times…