Return to hgm.mleFBByOptim.Rd CVS log

Up to [local] / OpenXM / src / R / r-packages / hgm / man

Annotation of OpenXM/src/R/r-packages/hgm/man/hgm.mleFBByOptim.Rd, Revision 1.4

1.4     ! takayama    1: % $OpenXM: OpenXM/src/R/r-packages/hgm/man/hgm.mleFBByOptim.Rd,v 1.3 2015/03/21 23:40:34 takayama Exp $
1.3       takayama    2: \name{hgm.z.mleFBByOptim}
1.4     ! takayama    3: \alias{hgm.z.mleFBByOptim}
1.3       takayama    4: %%Todo, write documents for hgm.z.mleDemo, hgm.ssFB
1.1       takayama    5: %\alias{hgm.ncso3}
                      6: %- Also NEED an '\alias' for EACH other topic documented here.
                      7: \title{
                      8:    MLE of Fisher-Bingham distribution by optim and HGM.
                      9: }
                     10: \description{
                     11:   It makes the maximal likelihood estimate (MLE) for the Fisher-Bingham
                     12:   distribution on S^d.
                     13: }
                     14: \usage{
1.4     ! takayama   15:  hgm.z.mleFBByOptim(d=0,ss=NULL,data=NULL,start=NULL,lb=NULL,ub=NULL,bigValue=10000)
1.1       takayama   16: }
                     17: %- maybe also 'usage' for other objects documented here.
                     18: \arguments{
                     19:   \item{d}{The dimension of the sphere}
                     20:   \item{ss}{Sufficient statistics}
                     21:   \item{data}{
                     22:      The argument data is a set of data on the d-dimensional sphere.
1.2       takayama   23:      Its format is an n by (d+1) matrix where n is the number of data.
                     24:      When data is given, ss must be NULL
1.1       takayama   25:     and ss is calculated from data by hgm.ssFB(data).
                     26:   }
                     27:   \item{start}{
                     28:      Starting point for the function optim. The default value is a random
                     29:      vector.
                     30:   }
                     31:   \item{lb}{
                     32:      An array of length sslen = (d+1)*(d+2)/2 + (d+1), each of which
                     33:      is the lower bound of the parameter. The default value is -100.
                     34:   }
                     35:   \item{ub}{
                     36:      An array of length sslen = (d+1)*(d+2)/2 + (d+1), each of which
                     37:      is the upper bound of the parameter. The default value is 100.
                     38:   }
                     39:   \item{bigValue}{
                     40:      It is used as a value wall to avoid that the evaluation point is out of
                     41:      the search domain defined by lb and ub.
                     42:   }
                     43: }
                     44: \details{
                     45:    It solves the MLE for the Fisher-Bingham distribution.
                     46:    The normalizing constant is evaluated by hgm_ko_ncfb (external program,
                     47:    which should in the path).
                     48:    The function
                     49: %  \code{\link[RCurl]{postForm}}
                     50:   \code{\link{optim}}
                     51:    is used for the optimization.
                     52:    The output is used as a starting point for the holonomic gradient method.
                     53:    See nk_fb_gen_c.rr of \url{http://www.math.kobe-u.ac.jp/Asir}.
1.2       takayama   54:    This function generates temporary work files whose names start with tmp.
                     55:    \code{data <- read.table(fileName,sep=",")} can be used to read CSV data
                     56:    from a file.
1.1       takayama   57: }
                     58: \value{
                     59: The output format is that of the function optim().
                     60: }
                     61: \references{
                     62: T. Koyama, H. Nakayama, K. Nishiyama, N. Takayama,
                     63: Holonomic Gradient Descent for the Fisher-Bingham Distribution on the d-dimensional Sphere,
                     64: Computational Statistics (2013)
                     65: \url{http://dx.doi.org/10.1007/s00180-013-0456-z}
                     66: }
                     67: \author{
                     68: T.Koyama, H.Nakayama, K.Nishiyama, N.Takayama.
                     69: }
                     70: \note{
                     71: %%  ~~further notes~~
                     72: }
                     73:
                     74: %% ~Make other sections like Warning with \section{Warning }{....} ~
                     75:
                     76: \seealso{
1.2       takayama   77: \code{\link{optim}}
1.1       takayama   78: }
                     79: \examples{
                     80: ## =====================================================
                     81: ## Example 1. Asteroid data in [N3OST2]
                     82: ## =====================================================
                     83: \dontrun{
                     84:   d <- 2
                     85:   ss <- c(0.3119,0.0292,0.0707,
                     86:                  0.3605,0.0462,
                     87:                            0.3276,
                     88:             0.0063,0.0054,0.0762)
                     89:   start <- c(0.1,0.1,1,1,1,-1,-1,-1,-1)
1.3       takayama   90:   hgm.z.mleFBByOptim(d=d,ss=ss,start=start)
1.1       takayama   91: }
                     92: }
                     93: % Add one or more standard keywords, see file 'KEYWORDS' in the
                     94: % R documentation directory.
                     95: \keyword{ Holonomic gradient method }
                     96: \keyword{ HGM }
                     97: \keyword{ Fisher-Bingham distribution on S^d}
                     98: \keyword{ MLE }
                     99: