nLL: Negative log likelihood of distances

In Rdistance: Density and Abundance from Distance-Sampling Surveys

Negative log likelihood of distances

Description

Return the negative log likelihood of observed detection distances given a likelihood and the estimated parameters.

Usage

nLL(a, ml, verbosity = 0)

Arguments

a	A vector of likelihood parameter values. Length and meaning depend on ml$series and ml$expansions. If no expansion terms were called for (i.e., ml$expansions = 0), the distance likelihood contain one or two canonical parameters (see Details). If one or more expansions are called for, coefficients for the expansion terms follow coefficients for the canonical parameters. i.e., length of this vector is (num Covars incl. intercept) + expansions + 1*(like %in% c("hazrate")).
ml	Either a Rdistance 'model frame' or an Rdistance 'fitted object'. Both are of class "dfunc". Rdistance 'model frames' are lists containing components necessary to estimate a distance function, but no estimates. Rdistance 'model frames' are typically produced by calls to parseModel. Rdistance 'fitted objects' are typically produced by calls to dfuncEstim. 'Fitted objects' are 'model frames' with additional components such as the parameters estimates, log likelihood value, convergence information, and the variance- covariance matrix of the parameters.
verbosity	The level of output produced during estimation. verbosity = 0 produces no output. Increasing values of verbosity >= 1 produce increasing levels of intermediate details. This is mostly used for de-bugging and checking convergence.

Details

Expansion Terms: If ml$expansions = k (k > 0), the expansion function specified by ml$series is called (see for example cosine.expansion). Assuming h_{ij}(x) is the j^{th} expansion term for the i^{th} distance and that c_1, c_2, \dots, c_k are (estimated) coefficients for the expansion terms, the likelihood contribution for the i^{th} distance is,

f(x|a,b,c_1,c_2,\dots,c_k) = f(x|a,b)(1 + \sum_{j=1}^{k} c_j h_{ij}(x)).

Value

A scalar, the negative of the log likelihood evaluated at parameters a.

See Also

See halfnorm.like and links there; dfuncEstim

Examples

# A halfnorm distance function 
fit <- dfuncEstim(sparrowDf, dist~1, likelihood = "halfnorm")
nLL(fit$par, fit, 3)
fit$loglik
ESW(fit)[1]

# Another way, b/c we have pnorm()
d <- distances(fit)
ones <- matrix(1, nrow = length(d), ncol = 1)
l <- halfnorm.like(fit$par, d, ones)
esw <-(pnorm(units::drop_units(fit$w.hi)
  , units::drop_units(fit$w.lo)
  , sd = exp(l$params)) - 0.5) * sqrt(2*pi) * exp(l$params)
-sum(log(l$L.unscaled/esw))

# A third way, b/c we have pnorm() and dnorm(). 
l2 <- dnorm(units::drop_units(d), mean = 0, sd = exp(fit$par))
scaler <- pnorm(units::drop_units(fit$w.hi), mean = 0, sd = exp(fit$par)) - 0.5 
-sum(log(l2/scaler))
 

Rdistance documentation built on Jan. 10, 2026, 1:07 a.m.