找回密码
 注册
查看: 305|回复: 0

R语言 rsae包 fitsaemodel()函数中文帮助文档(中英文对照)

[复制链接]
发表于 2012-9-28 21:15:23 | 显示全部楼层 |阅读模式
fitsaemodel(rsae)
fitsaemodel()所属R语言包:rsae

                                         Fit SAE model using various methods
                                         适合SAE模型,用不同的方法

                                         译者:生物统计家园网 机器人LoveR

描述----------Description----------

fitsaemodel is the workhorse function. It estimates SAE models that have been set up by saemodel (or synthetic data generated by makedata) by various (robust) estimation methods.
fitsaemodel是主力功能。据估计,SAE模式,已成立由saemodel(或合成makedata)由各种(强大)的估计方法所产生的数据。


用法----------Usage----------


fitsaemodel(method, model, ...)

convergence(object)

## S3 method for class 'fitsaemodel'
print(x, digits=6, ...)
## S3 method for class 'fitsaemodel'
summary(object, digits=6, ...)
## S3 method for class 'fitsaemodel'
coef(object, type="both", ...)



参数----------Arguments----------

参数:method
character string defining the method to be used; currently, either method="ml" for (non-robust) maximum likelihood or method="huberm" for Huber-type M-estimation  
字符的字符串,定义要使用的方法,目前,无论是method="ml"(非强大)最大的可能性或method="huberm"贝尔型M-估计


参数:model
a "saemodel" object (i.e., a SAE model; see saemodel)  
"saemodel"对象(即SAE模型;saemodel)


参数:x
used by the print method  
使用print方法


参数:digits
used by the print and summary methods; number of decimal places to be shown  
使用print和summary方法;要显示的小数位数的数字


参数:object
an object of the class "fitsaemodel"; i.e., a fitted model  
一个对象的类"fitsaemodel",也就是说,一个合适的模型


参数:type
character string use in the coef method; it can take one of the following possibilities: "both", "ranef", or "fixef". The first reports both, random and fixed effects (default).   
字符串中使用的coef方法,它可以采取以下几种可能性之一:"both","ranef"或"fixef"。第一次报告,随机和固定效应(默认值)。


参数:...
additional arguments delivered to either fitsaemodel.control   
提供额外的参数为fitsaemodel.control


Details

详细信息----------Details----------

The function fitsaemodel is a wrapper function that calls the algorithm associated with a particular method. Two methods are currently implemented
函数fitsaemodel是一个包装函数调用的算法与一个特定的方法。目前,有两种方法实现

maximum likelihood (method="ml"),
最大似然(method="ml"),

Huber-type M-estimation (method="huberm"; cf. RML II of Richardson and Welsh, 1995).
胡贝尔型M-估计(method="huberm";比照RML II理查德森和威尔士,1995年)。

The call for ML is straightforward: fitsaemodel(method="ml", model), where model is a SAE model generated by saemodel. Note that ML is not a robust fitting method.
ML的要求很简单:fitsaemodel(method="ml", model),其中model是一个SAE模型所产生的saemodel。请注意,ML是不是一个强大的拟合方法。

The call for Huber-type M-estimaton (with Huber psi-function) is: fitsaemodel(method="huberm", model, k), where model is a SAE model generated by saemodel, and k is the robustness tuning constant of the Huber psi-function.
胡贝尔型M-estimaton(胡贝尔PSI功能)的要求是:fitsaemodel(method="huberm", model, k),这里model是一个SAE模型所产生的saemodel,k是鲁棒性调整的Huber PSI的功能保持不变。

By default, the "huberm" method is initialized by means of pre-determined robust estimates of a fixed-effects model (centered by the median instead of the mean); see Schoch (2012) for the details.
默认情况下,"huberm"方法是通过预先确定的固定效应模型为中心的中位数,而不是平均地估计初始化的详细信息,请参阅肖赫(2012年)。

If your data are supposed to be heavily contaminated (or if the default algorithm did not converge), you may initialize the fitsaemodel alogrithm with a high-breakdown-point estimate. The rsae package offers two methods to initialize the algorithm, "lts" and "s"; see below. NOTE, you have to install the robustbase package in order to use these methods. The initialization methods are called in the fitsaemodel device (as additional argument), using
如果您的数据都被认为是严重污染的(或默认的算法不收敛),你可能会初始化fitsaemodelalogrithm的具有高击穿点估计。 rsae包提供了两种方法来初始化算法,"lts"的和"s"见下文。请注意,您必须安装robustbase包,以便使用这些方法。 fitsaemodel设备(如额外的参数)的初始化方法被调用,使用

init="lts", for fast-LTS regression form robustbase; see also Rousseeuw and Van Driessen (2006),
init="lts",快速LTS回归的形式robustbase;也看到Rousseeuw和Van Driessen的(2006年),

init="s", for a regression S-estimator from robustbase; see also Maronna et al. (2006).
init="s",robustbase;也看到Maronna的等的回归S-估计。 (2006年)。

For more details on the methods, you are refered to the documentation of robustbase. In general, for small to medium datasets, both methods are equivalent. For data with more than 50,000 observations, the S-estimator is considerably faster. (If the "ml" does not converge, you may initialize it analogously–though, it may be rather inefficient.)
的方法的详细信息,参考,以您的文档robustbase。在一般情况下,对于小到中等数据集,这两种方法是等效的。对于具有超过50,000个观测值的数据,在S-估计是相当快。 (如果"ml"不衔接,可以初始化类似,虽然,它可能是非常低效的。)

For both method="ml" and method="huberm", the estimates are obtained by means of a nested loop of IRWLS approaches and Brent's zeroin method (Brent, 1973). All the functions/subroutines are optimized to be rich in BLAS level-one operations (Blackford et al., 2002) and draw heavily on LAPACK (Anderson et al., 2000).
对于两个method="ml"和method="huberm",估计是一个嵌套循环的IRWLS方法和布伦特的zeroin方法(布伦特,1973年)获得通过。所有的功能/子程序进行了优化,含有丰富的,BLAS级操作(布莱克福德等,2002),并大量借鉴LAPACK(Anderson等,2000)。


值----------Value----------

An instance of the class "fitmodel"
实例的类"fitmodel",


(作者)----------Author(s)----------



Tobias Schoch




参考文献----------References----------









参见----------See Also----------

fitsaemodel.control  
fitsaemodel.control


实例----------Examples----------


#generate the synthetic data/model[生成的合成数据/模型]
mymodel <- makedata()
#compute Huber M-estimation type estimates of the model "mymodel"[计算胡贝尔M-估计估计模式“mymodel”]
#robustness tuning constant k = 2[鲁棒性调整常数k = 2]
myfittedmodel <- fitsaemodel("huberm", mymodel, k=2)
myfittedmodel
#get a summary of the model[的模型中得到的摘要]
summary(myfittedmodel)

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。
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 注册

本版积分规则

手机版|小黑屋|生物统计家园 网站价格

GMT+8, 2024-11-27 03:13 , Processed in 0.025177 second(s), 16 queries .

Powered by Discuz! X3.5

© 2001-2024 Discuz! Team.

快速回复 返回顶部 返回列表