Prints a detailed summary of the fitted model, including coefficient
estimates with robust standard errors, z- or t-values, p-values, and
goodness-of-fit statistics (log-likelihood, AIC, BIC, deviance when
available). The output concludes with estimated population sizes
\(\hat{\xi} = \sum_i N_i^{\hat{\alpha}_i}\)
for each covariate group defined by cov_alpha, along with
bias-corrected estimates and confidence intervals computed by
popsize.
Usage
# S3 method for class 'uncounted'
summary(object, total = FALSE, ...)Arguments
- object
An
"uncounted"object returned byestimate_hidden_pop.- total
Logical; if
TRUE, include a total row in the population-size table. DefaultFALSE.- ...
Additional arguments (currently ignored).
Value
Invisibly returns the coefficient summary matrix (estimates, standard errors, test statistics, and p-values).
Details
For OLS and NLS fits, inference uses t-statistics with
\(n - p\) degrees of freedom. For Poisson and NB fits, z-statistics
(normal approximation) are used. Standard errors are always
HC-robust (type controlled by vcov_type in the original call).