R/F_NBpsiInfl.R

In RCM: Fit row-column association models with the negative binomial distribution for the microbiome

#' The influence function for the psis
#'
#' @param rcm an rcm object
#' @param Dim the required dimensions
#'
#' @return The influence of every single observation
#'  on the psi value of this dimension
NBpsiInfl = function(rcm, Dim = 1) {
    mu = extractE(rcm, seq_len(Dim))
    # Take also lower dimensions into account
    # here
    thetaMat = matrix(byrow = TRUE, nrow = nrow(rcm$X),
        ncol = ncol(rcm$X), data = rcm$thetas[,
            switch(as.character(Dim), `0` = "Independence",
                `0.5` = "Filtered", paste0("Dim",
                Dim))])
    reg = if (is.null(rcm$covariates)) {
        rcm$rMat[, Dim, drop = FALSE] %*% rcm$cMat[Dim,, drop = FALSE]
    } else {getRowMat(sampleScore = rcm$covariates %*% rcm$alpha[, Dim],
                      responseFun = rcm$responseFun,
        NB_params = rcm$NB_params[, , Dim])}
    rcm$X = correctXMissingness(rcm$X, mu, rcm$NApresent, naId = is.na(rcm$X))

    -((rcm$X - mu) * (thetaMat + mu))/(reg *
        (thetaMat + rcm$X) * mu)
}