Bivariate Zero-Inflated Count Models Using Copulas

Maximum likelihood estimation of copula-based zero-inflated (and non-inflated) Poisson and negative binomial count models, based on the article . Supports Frank and Gaussian copulas. Allows for mixed margins (e.g., one margin Poisson, the other zero-inflated negative binomial), and several marginal link functions. Built-in methods for publication-quality tables using 'texreg', post-estimation diagnostics using 'DHARMa', and testing for marginal zero-modification via . For information on copula regression for count data, see Genest and Nešlehová (2007) as well as Nikoloulopoulos (2013) . For information on zero-inflated count regression generally, see Lambert (1992) < https://www.jstor.org/stable/1269547>. The author acknowledges support by NSF DMS-1925119 and DMS-212324.

bizicount

The bizicount R-package is primarily for estimating copula-based bivariate count regression models, with and without zero-inflation or overdispersion. However, its full suite of functions can:

Estimate copula-based bivariate zero-inflated count regression models, as well as non-inflated models.
Estimate univariate zero-inflated count models (zic.reg)
Carry out post estimation diagnostics using the DHARMa package (simulate.bizicount, simulate.zicreg, make_DHARMa)
Produce professional tables in latex, word, or plain-text using the texreg package (extract.bizicount, extract.zicreg).
Test for zero modification using zi_test() (He et al. 2019).
Evaluate univariate zero-inflated count distribution CDFs, PDFs, and quantile functions, and generate random zero-inflated counts (pzip or pzinb, dzip or dzinb, qzip or qzinb, and rzip or rzinb)