logLik: Log-Likelihood

In countprop: Calculate Model-Based Metrics of Proportionality on Count-Based Compositional Data

Log-Likelihood

Description

Calculates the log-likelihood, under the multinomial logit-Normal model.

Usage

logLik(v, y, ni, S, invSigma)

Arguments

v	The additive log-ratio transform of y
y	Compositional dataset
ni	The row sums of y
S	Covariance of v
invSigma	The inverse of the Sigma matrix

Value

The estimated log-likelihood under the Multinomial logit-Normal distribution.

Examples

data(singlecell)
mle.sim <- mlePath(singlecell, tol=1e-4, tol.nr=1e-4, n.lambda = 2, n.cores = 1)

n <- NROW(singlecell)


logLik(mle.sim$est.min$v,
      singlecell,
      n,
      cov(mle.sim$est.min$v),
      mle.sim$est.min$Sigma.inv)

countprop documentation built on Aug. 18, 2023, 9:06 a.m.