negative.binomial.Rd
Specifies the information required to fit a Negative Binomial generalized
linear model, with known theta
parameter, using glm()
.
negative.binomial(theta = stop("'theta' must be specified"), link = "log")
The known value of the additional parameter, theta
.
The link function, as a character string, name or one-element
character vector specifying one of log
, sqrt
or identity
, or an object of class
"link-glm"
.
An object of class "family"
, a list of functions and
expressions needed by glm()
to fit a Negative Binomial
generalized linear model.
Venables, W. N. and Ripley, B. D. (2002) Modern Applied Statistics with S-PLUS. Fourth Edition. Springer.
# Fitting a Negative Binomial model to the quine data
# with theta = 2 assumed known.
#
glm(Days ~ .^4, family = negative.binomial(2), data = quine)
#>
#> Call: glm(formula = Days ~ .^4, family = negative.binomial(2), data = quine)
#>
#> Coefficients:
#> (Intercept) EthN SexM
#> 3.0564 -0.1386 -0.4914
#> AgeF1 AgeF2 AgeF3
#> -0.6227 -2.3632 -0.3784
#> LrnSL EthN:SexM EthN:AgeF1
#> -1.9577 -0.7524 0.1029
#> EthN:AgeF2 EthN:AgeF3 EthN:LrnSL
#> -0.5546 0.0633 2.2588
#> SexM:AgeF1 SexM:AgeF2 SexM:AgeF3
#> 0.4092 3.1098 1.1145
#> SexM:LrnSL AgeF1:LrnSL AgeF2:LrnSL
#> 1.5900 2.6421 4.8585
#> AgeF3:LrnSL EthN:SexM:AgeF1 EthN:SexM:AgeF2
#> NA -0.3105 0.3469
#> EthN:SexM:AgeF3 EthN:SexM:LrnSL EthN:AgeF1:LrnSL
#> 0.8329 -0.1639 -3.5493
#> EthN:AgeF2:LrnSL EthN:AgeF3:LrnSL SexM:AgeF1:LrnSL
#> -3.3315 NA -2.4285
#> SexM:AgeF2:LrnSL SexM:AgeF3:LrnSL EthN:SexM:AgeF1:LrnSL
#> -4.1914 NA 2.1711
#> EthN:SexM:AgeF2:LrnSL EthN:SexM:AgeF3:LrnSL
#> 2.1029 NA
#>
#> Degrees of Freedom: 145 Total (i.e. Null); 118 Residual
#> Null Deviance: 280.2
#> Residual Deviance: 172 AIC: 1095