cl_l

theta

a vector of observation for the subject.

covariate for the particular subject

corresponding distance of selected pair

dwdv

combination of the pairs included into the composite likelihood

cmwdv

number of parameter (i.e. length of theta)

identity matrix with dimension <code>J+1</code>

base

number of category among (ALL) observed response.

number of covariate (i.e. number of column of <code>X</code>)

<code>cl_l</code> Calculate the negative composite log-likelihood value for a particular subject given parameter value and other input variables

Composite likelihood parameter estimate and asymptotic covariance matrix are calculated for the spatial ordinal data with replications, where spatial ordinal response with covariate and both spatial exponential covariance within subject and independent and identically distributed measurement error. Parameter estimation can be performed by either solving the gradient function or maximizing composite log-likelihood. Parametric bootstrapping is used to estimate the Godambe information matrix and hence the asymptotic standard error and covariance matrix with parallel processing option. Moreover, the proposed surrogate residual, which extends the results of Liu and Zhang (2017) <doi: 10.1080/01621459.2017.1292915>, can act as a useful tool for model diagnostics.

cl_l: Composite Likelihood Calculation for Spatial Ordinal Data without Replications (for implmentation)

Description

Usage

Arguments

Value