cmp_cov: Computes the covariance between the samples in the Markov...
In khabbazian/rdssim: A simple respondent-driven sampling simulator

Computes the covariance between the samples in the Markov chain/tree using eigenvalues and eigenvectors of the Laplacian matrix of the social network.

1	cmp_cov(A, covariates, distance = 1)

`A`	adjacency matrix of the undirected network.
`covariates`	the characteristic of interest for each node in the network.
`distance`	distance between the random variables in the Markov chain/tree.

Let \pi, y, and d be the stationary distribution, the covariate and the distance respectively. Then cov(y(X_i),y(X_i+d)) = \sum_i=2^N <y,f_i>^2_\pi \lambda_i^d.

khabbazian/rdssim documentation built on May 20, 2019, 9:22 a.m.