1d:T782,# NBC: maximum likelihood linear negative binomial regression ml.nbc is a maximum likelihood function for estimating canonical linear negative binomial (NB-C) data. ```r ml.nbc(formula, data, start=NULL, verbose=FALSE) ``` ## Arguments - `formula`: an object of class '"formula"': a symbolic description of the model to be fitted. The details of model specification are given under 'Details'. - `data`: a mandatory data frame containing the variables in the model. - `start`: an optional vector of starting values for the parameters. - `verbose`: a logical flag to indicate whether the fit information should be printed. ## Details ml.nbc is used like glm.nb, but without saving ancillary statistics. ## Returns The function returns a dataframe with the following components: - **Estimate**: ML estimate of the parameter - **SE**: Asymptotic estimate of the standard error of the estimate of the parameter - **Z**: The Z statistic of the asymptotic hypothesis test that the population value for the parameter is 0. - **LCL**: Lower 95% confidence interval for the parameter estimate. - **UCL**: Upper 95% confidence interval for the parameter estimate. ## References Hilbe, J.M. (2011), Negative Binomial Regression, second edition, Cambridge University Press. ## Author(s) Andrew Robinson, Universty of Melbourne, Australia, and Joseph M. Hilbe, Arizona State University, and Jet Propulsion Laboratory, California Institute of Technology ## See Also `glm.nb`, `ml.nb1`, `ml.nb2` ## Examples ```r # Table 10.12, Hilbe. J.M. (2011), Negative Binomial Regression, # 2nd ed. Cambridge University Press (adapted) ## Not run: data(medpar) nobs <- 50000 x2 <- runif(nobs) x1 <- runif(nobs) xb <- 1.25*x1 + .1*x2 - 1.5 mu <- 1/(exp(-xb)-1) p <- 1/(1+mu) r <- 1 gcy <- rnbinom(nobs, size=r, prob = p) test <- data.frame(gcy, x1, x2) nbc <- ml.nbc(gcy ~ x1 + x2, data=test) nbc ## End(Not run) ```

ml.nbc function