inst/doc/using-maxlik.Rnw

29 \newcommand{\maxlik}{\texttt{maxLik}\xspace}
38 % \VignetteIndexEntry{Maximum likelihood estimation with maxLik}
40 \title{Maximum Likelihood Estimation with \emph{maxLik}}
47 library(maxLik)
79 \subsection{The maxLik function}
83 the same name, \verb|maxLik|.  It is a wrapper around the underlying
86 can use the convenience methods, such as \verb|summary| or
87 \verb|logLik|.  It is important to keep in mind that \maxlik
91 log-likelihood function (argument \verb|logLik|) and the start value
105 m <- maxLik(loglik, start=c(mu=1, sigma=2))
107 summary(m)
117 This example demonstrates a number of key features of \verb|maxLik|:
129   components are named, those names are also displayed in \verb|summary|
131 \item \verb|summary| method displays a handy summary of the results,
134 \item \verb|maxLik| (and other auxiliary optimizers in the package) is a
137 As we did not specify the optimizer, \verb|maxLik| picked
143 Besides summary, \verb|maxLik| also contains a number of utility
154 \item Other functions include \verb|logLik| to return the
169 numeric gradient \verb|maxLik| computed internally
177 show how to supply gradient to the \verb|maxLik| function.
215 (remember, \verb|maxLik| is a maximizer not minimizer)
230 m <- maxLik(negSSE, start=c(0,0,0))
232 summary(m, eigentol=1e-15)
236 1, 1)$.  We have amended summary with an extra argument,
239 see the documentation of \verb|summary.maxLik|.  It makes no
261 We can add gradient to \verb|maxLik| as an additional argument
264 m <- maxLik(negSSE, grad=grad, start=c(0,0,0))
265 summary(m, eigentol=1e-15)
286 m <- maxLik(negSSE, grad=grad, hess=hess, start=c(0,0,0))
287 summary(m, eigentol=1e-15)
318 m <- maxLik(negSSEA, start=c(0,0,0))
319 summary(m, eigentol=1e-15)
380 m <- maxLik(loglik, start=c(mu=1, sigma=2),
382 summary(m)
504 m <- maxLik(loglik, gradlikB, start=c(mu=1, sigma=2),
506 summary(m)
523 m <- maxLik(loglikB, start=c(mu=1, sigma=2),
525 summary(m)
558 m <- maxLik(loglikB, start=c(mu=1, sigma=2),
561 summary(m)
563 The first option, \verb|printLevel=3|, make \verb|maxLik| to print
602 m <- maxLik(loglikB, start=c(mu=1, sigma=2),
605 summary(m)
635 \verb|maxLik| function.  For instance, let's change the
645 m <- maxLik(loglik, start=c(mu=1, sigma=2), x=x)
648 summary(m)
651 \verb|maxLik|'s arguments' names.  If that happens, it
658 \verb|maxLik| function is basically a wrapper around a number of
689 summary(m)
691 Note that the summary output is slightly different: it reports the
694 return an object of \emph{maxim}-class, not \emph{maxLik}-class.
717 m <- maxLik(negSSEA, start=c(x1=0, x2=0, x3=0))
720 summary(m)
779 m <- maxLik(loglik, start=c(x1=0, x2=0, x3=0))
780 summary(m)
822 m <- maxLik(loglik, start=c(mu=1, sigma=1),
825 summary(m)
874 summary(m)
913 summary(m)
952 summary(m)