lillardhaz: Simultaneous-Equations Hazard/Probit Models with a
Gaussian-Copula Correlation
Fits two-equation simultaneous models in the style of
Lillard (1993) <doi:10.1016/0304-4076(93)90106-F>: a pair of processes,
each either a binary probit outcome or a continuous-time hazard
duration, linked through a single correlation parameter between their
underlying error terms and estimated jointly by maximum likelihood via
a Gaussian copula. Equation 1 may be a probit, a log-normal
accelerated-failure-time hazard, or a piecewise (linear-log-hazard)
Gompertz hazard with an arbitrary, user-specified number of segments;
equation 2 is always a hazard, log-normal or piecewise Gompertz. Models
can also be fit under independence (correlation fixed at zero), which
reduces exactly to two separate univariate fits. See the package
vignette/manual for the full likelihood derivation and a
simulation-based verification.
Documentation:
Downloads:
Linking:
Please use the canonical form
https://CRAN.R-project.org/package=lillardhaz
to link to this page.