nix2_marginal_lhood: Marginal likelihood function
In kleinschmidt/beliefupdatr: Belief updating for phonetic adaptation

Compute the marginal likelihood of data given NIX^2 prior, or p(x) = \int \mathrm d σ_2 \mathrm dμ p(x | μ, σ^2) p(μ, σ^2).

1 2	nix2_marginal_lhood(p, x = NA, n = length(x), xbar = mean(x), ss = s2 * (n - 1), s2 = var(x), log = FALSE)

`p`	named list of distribution parameters (mu, sigma2, nu, and kappa)
`x`	vector of observations to update with (optional if summary stats are passed)
`n`	number of observations (overrides x)
`xbar`	observed mean (overrides x)
`ss`	observed sum of squares (overrides x and s2)
`s2`	observed sample variance (ss / (n-1); overrides x)
`log`	Default FALSE.