📄 glm.nb.html
字号:
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html><head><title>R: Fit a Negative Binomial Generalized Linear Model</title>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../../R.css">
</head><body>
<table width="100%" summary="page for glm.nb {MASS}"><tr><td>glm.nb {MASS}</td><td align="right">R Documentation</td></tr></table>
<h2>Fit a Negative Binomial Generalized Linear Model</h2>
<h3>Description</h3>
<p>
A modification of the system function <code><a href="../../VGAM/html/notdocumentedyet.html">glm</a>()</code> to include
estimation of the additional parameter, <code>theta</code>, for a
Negative Binomial generalized linear model.
</p>
<h3>Usage</h3>
<pre>
glm.nb(formula, data, weights, subset, na.action,
start = NULL, etastart, mustart,
control = glm.control(...), method = "glm.fit",
model = TRUE, x = FALSE, y = TRUE, contrasts = NULL, ...,
init.theta, link = log)
</pre>
<h3>Arguments</h3>
<table summary="R argblock">
<tr valign="top"><td><code>formula, data, weights, subset, na.action, start, etastart,
mustart, control, method, model, x, y, contrasts, ...</code></td>
<td>
arguments for the <code><a href="../../VGAM/html/notdocumentedyet.html">glm</a>()</code> function.
Note that these exclude <code>family</code> and <code>offset</code>
(but <code><a href="../../stats/html/offset.html">offset</a>()</code> can be used).
</td></tr>
<tr valign="top"><td><code>init.theta</code></td>
<td>
Optional initial value for the theta parameter. If omitted a moment
estimator after an initial fit using a Poisson GLM is used.
</td></tr>
<tr valign="top"><td><code>link</code></td>
<td>
The link function. Currently must be one of <code>log</code>, <code>sqrt</code>
or <code>identity</code>.
</td></tr>
</table>
<h3>Details</h3>
<p>
An alternating iteration process is used. For given <code>theta</code> the GLM
is fitted using the same process as used by <code>glm()</code>. For fixed means
the <code>theta</code> parameter is estimated using score and information
iterations. The two are alternated until convergence of both. (The
number of alternations and the number of iterations when estimating
<code>theta</code> are controlled by the <code>maxit</code> parameter of
<code>glm.control</code>.)
</p>
<p>
Setting <code>trace > 0</code> traces the alternating iteration
process. Setting <code>trace > 1</code> traces the <code>glm</code> fit, and
setting <code>trace > 2</code> traces the estimation of <code>theta</code>.
</p>
<h3>Value</h3>
<p>
A fitted model object of class <code>negbin</code> inheriting from <code>glm</code>
and <code>lm</code>. The object is like the output of <code>glm</code> but contains
three additional components, namely <code>theta</code> for the ML estimate of
theta, <code>SE.theta</code> for its approximate standard error (using
observed rather than expected information), and <code>twologlik</code> for
twice the log-likelihood function.</p>
<h3>References</h3>
<p>
Venables, W. N. and Ripley, B. D. (2002)
<EM>Modern Applied Statistics with S.</EM> Fourth edition. Springer.
</p>
<h3>See Also</h3>
<p>
<code><a href="../../VGAM/html/notdocumentedyet.html">glm</a></code>, <code><a href="negative.binomial.html">negative.binomial</a></code>,
<code><a href="anova.negbin.html">anova.negbin</a></code>, <code><a href="summary.negbin.html">summary.negbin</a></code>,
<code><a href="theta.md.html">theta.md</a></code>
</p>
<h3>Examples</h3>
<pre>
quine.nb1 <- glm.nb(Days ~ Sex/(Age + Eth*Lrn), data = quine)
quine.nb2 <- update(quine.nb1, . ~ . + Sex:Age:Lrn)
quine.nb3 <- update(quine.nb2, Days ~ .^4)
anova(quine.nb1, quine.nb2, quine.nb3)
</pre>
<hr><div align="center">[Package <em>MASS</em> version 7.2-44 <a href="00Index.html">Index]</a></div>
</body></html>
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -