knockoffDMC.RdThis function constructs knockoffs of variables distributed as a discrete Markov chain.
knockoffDMC(X, pInit, Q, groups = NULL, seed = 123, cluster = NULL, display_progress = FALSE)
| X | an integer matrix of size n-by-p containing the original variables. |
|---|---|
| pInit | an array of length K, containing the marginal distribution of the states for the first variable. |
| Q | an array of size (p-1,K,K), containing a list of p-1 transition matrices between the K states of the Markov chain. |
| groups | an array of length p, describing the group membership of each variable (default: NULL). |
| seed | an integer random seed (default: 123). |
| cluster | a computing cluster object created by makeCluster (default: NULL). |
| display_progress | whether to show progress bar (default: FALSE). |
An integer matrix of size n-by-p containing the knockoff variables.
Each element of the matrix X should be an integer value between 0 and K-1. The transition matrices contained in Q are defined such that \(P[X_{j+1}=k|X_{j}=l]=Q[j,l,k]\).
Sesia M, Sabatti C, Candès EJ (2019). “Gene hunting with hidden Markov model knockoffs.” Biometrika, 106, 1--18. doi: 10.1093/biomet/asy033 . Sesia M, Katsevich E, Bates S, Candès E, Sabatti C (2019). “Multi-resolution localization of causal variants across the genome.” bioRxiv. doi: 10.1101/631390 .
Other knockoffs: knockoffGenotypes,
knockoffHMM,
knockoffHaplotypes
# Generate data p = 10; K = 5; pInit = rep(1/K,K) Q = array(stats::runif((p-1)*K*K),c(p-1,K,K)) for(j in 1:(p-1)) { Q[j,,] = Q[j,,] / rowSums(Q[j,,]) } X = sampleDMC(pInit, Q, n=20) # Generate knockoffs Xk = knockoffDMC(X, pInit, Q) # Generate group-knockoffs for groups of size 3 groups = rep(seq(p), each=3, length.out=p) Xk = knockoffDMC(X, pInit, Q, groups=groups)