wle.gamma(wle)
wle.gamma()所属R语言包:wle
Robust Estimation in the Gamma model
在Gamma模型的鲁棒估计
译者:生物统计家园网 机器人LoveR
描述----------Description----------
wle.gamma is used to robust estimate the shape and the scale parameters via Weighted Likelihood, when the majority of the data are from a gamma distribution.
wle.gamma稳健估计的形状和尺度参数,通过加权似然,当大多数的数据是从伽玛分布。
用法----------Usage----------
wle.gamma(x, boot=30, group, num.sol=1, raf="HD", smooth=0.008,
tol=10^(-6), equal=10^(-3), max.iter=500,
shape.int=c(0.01, 100), use.smooth=TRUE, tol.int,
verbose=FALSE, maxiter=1000)
参数----------Arguments----------
参数:x
a vector contain the observations.
一个向量包含的意见。
参数:boot
the number of starting points based on boostrap subsamples to use in the search of the roots.
基于自举子样本的起点,使用在搜索的根的数目。
参数:group
the dimension of the bootstap subsamples. The default value is max(round(size/4),var) where size is the number of observations and var is the number of variables.
的维度的bootstap子样本。默认值是max(round(size/4),var)size的一些意见和var是变量的数目。
参数:num.sol
maximum number of roots to be searched.
要搜索的最大根数。
参数:raf
type of Residual adjustment function to be use:
类型的残余调节功能,可以使用:
raf="HD": Hellinger Distance RAF,
raf="HD":Hellinger距离RAF,
raf="NED": Negative Exponential Disparity RAF,
raf="NED":负指数差异RAF,
raf="SCHI2": Symmetric Chi-Squared Disparity RAF.
raf="SCHI2":对称卡方差异皇家空军。
参数:smooth
the value of the smoothing parameter.
的平滑化参数的值。
参数:tol
the absolute accuracy to be used to achieve convergence of the algorithm.
要使用的绝对精度实现算法的收敛性。
参数:equal
the absolute value for which two roots are considered the same. (This parameter must be greater than tol).
绝对的值,两个根被认为是相同的。 (此参数必须大于tol)。
参数:max.iter
maximum number of iterations for the main function.
为主要功能的迭代的最大数量。
参数:shape.int
a 2 dimension vector for the interval search of the shape parameter.
2维矢量形状参数的区间。
参数:use.smooth
if FALSE the unsmoothed model is used. This is usefull when the integration routine does not work well.
如果FALSE未滤波的模型使用。这是有用的,当积分程序无法正常工作。
参数:tol.int
the absolute accuracy to be used in the integration routine. The default value is tol*10^{-4}.
整合例程中要使用的绝对精度。默认值是tol*10^{-4}。
参数:verbose
if TRUE warnings are printed.
如果TRUE警告被打印出来。
参数:maxiter
maximum number of iterations. This value is passed to uniroot function.
最大迭代次数。该值被传递到uniroot函数。
Details
详细信息----------Details----------
The gamma is parametrized as follows (α = scale, ω = shape):
伽马参数化如下(α = scale,ω = shape):
f(x) = 1/(α^ω Gamma(ω)) x^(ω-1) e^-(x/α)
f(x) = 1/(α^ω Gamma(ω)) x^(ω-1) e^-(x/α)
for x > 0, α > 0 and ω > 0.
x > 0,α > 0和ω > 0。
The function use uniroot to solve the estimating equation for shape, errors from uniroot are handled by try. If errors occurs then the function returns NA.
功能使用uniroot解决的估算公式为shape“的错误uniroot是处理try。如果发生错误,则函数返回NA。
You can use shape.int to avoid them. It also use a fortran routine (dqagp) to calculate the smoothed model, i.e., evaluate the integral. Sometime the accuracy is not satisfactory, you can use use.smooth=FALSE to have an approximate estimation using the model instead of the smoothed model.
您可以使用shape.int,以避免他们。它也可以使用一个Fortran的程序(dqagp)的计算平滑的模式,即,评估的积分。有时的准确性不理想,你可以使用use.smooth=FALSE有一个近似估计的模型,而不是平滑的模型。
The Folded Normal distribution is use as kernel. The bandwith is smooth*shape/scale^2.
倍数正态分布为核心。的带宽是smooth*shape/scale^2。
值----------Value----------
wle.gamma returns an object of class "wle.gamma".
wle.gamma返回一个对象的class"wle.gamma"的。
Only print method is implemented for this class.
只打印的方法来实现这个类。
The object returned by wle.gamma are:
对象返回wle.gamma是:
<table summary="R valueblock"> <tr valign="top"><td>shape</td> <td> the estimator of the shape parameter, one value for each root found.</td></tr> <tr valign="top"><td>scale</td> <td> the estimator of the scale parameter, one value for each root found.</td></tr> <tr valign="top"><td>rate</td> <td> the estimator of the rate parameter (1/scale), one value for each root found.</td></tr> <tr valign="top"><td>tot.weights</td> <td> the sum of the weights divide by the number of observations, one value for each root found.</td></tr> <tr valign="top"><td>weights</td> <td> the weights associated to each observation, one column vector for each root found.</td></tr> <tr valign="top"><td>f.density</td> <td> the non-parametric density estimation.</td></tr> <tr valign="top"><td>m.density</td> <td> the smoothed model.</td></tr> <tr valign="top"><td>delta</td> <td> the Pearson residuals.</td></tr> <tr valign="top"><td>call</td> <td> the match.call().</td></tr> <tr valign="top"><td>tot.sol</td> <td> the number of solutions found.</td></tr> <tr valign="top"><td>not.conv</td> <td> the number of starting points that does not converge after the max.iter iteration are reached.</td></tr>
<table summary="R valueblock"> <tr valign="top"> <TD> shape</ TD> <TD>的形状参数的估计,每一根发现。</ TD> </ TR> <tr valign="top"> <TD> scale</ TD> <TD>的尺度参数的估计,每根发现的一个值。</ TD> </ TR> < TR VALIGN =“顶”> <TD>rate </ TD> <TD>估计的速度的参数(1/scale),一个值每根发现。</ TD> </ TR> < TR VALIGN =“顶”> <TD>tot.weights </ TD> <TD>的权重总和除以观测值的数量,每根发现的一个值。</ TD> </ TR> < TR VALIGN =“顶”> <TD>weights </ TD> <TD>相关的权重给每个观察,每一根发现一个列向量。</ TD> </ TR> <TR VALIGN =“顶“<TD> f.density </ TD> <TD>非参数密度估计。</ TD> </ TR> <tr valign="top"> <TD>m.density / TD> <TD>的平滑模型。</ TD> </ TR> <tr valign="top"> <TD>delta </ TD> <TD> Pearson残差。</ TD> < / TR> <tr valign="top"> <TD> call </ TD> <TD> match.call()。</ TD> </ TR> <tr valign="top"> < tot.sol TD> </ TD> <TD>找到解决方案的数量。</ TD> </ TR> <tr valign="top"> <TD>not.conv </ TD> <TD >起点后max.iter迭代不收敛的数目达到。</ TD> </ TR>
</table>
</ TABLE>
(作者)----------Author(s)----------
Claudio Agostinelli
参考文献----------References----------
Markatou, M., Basu, A. and Lindsay, B.G., (1998). Weighted likelihood estimating equations with a bootstrap root search, Journal of the American Statistical Association, 93, 740-750.
实例----------Examples----------
library(wle)
x <- rgamma(n=100, shape=2, scale=2)
wle.gamma(x)
x <- c(rgamma(n=30, shape=2, scale=2), rgamma(n=100, shape=20, scale=20))
wle.gamma(x, boot=10, group=10, num.sol=2) # depending on the sample, one or two roots. [取决于样品,一个或两个根。]
转载请注明:出自 生物统计家园网(http://www.biostatistic.net)。
注:
注1:为了方便大家学习,本文档为生物统计家园网机器人LoveR翻译而成,仅供个人R语言学习参考使用,生物统计家园保留版权。
注2:由于是机器人自动翻译,难免有不准确之处,使用时仔细对照中、英文内容进行反复理解,可以帮助R语言的学习。
注3:如遇到不准确之处,请在本贴的后面进行回帖,我们会逐渐进行修订。
|