fastbw(rms)
fastbw()所属R语言包:rms
Fast Backward Variable Selection
快退变量选择
译者:生物统计家园网 机器人LoveR
描述----------Description----------
Performs a slightly inefficient but numerically stable version of fast backward elimination on factors, using a method based on Lawless and Singhal (1978). This method uses the fitted complete model and computes approximate Wald statistics by computing conditional (restricted) maximum likelihood estimates assuming multivariate normality of estimates. fastbw deletes factors, not columns of the design matrix. Factors requiring multiple d.f. will be retained or dropped as a group. The function prints the deletion statistics for each variable in turn, and prints approximate parameter estimates for the model after deleting variables. The approximation is better when the number of factors deleted is not large. For ols, the approximation is exact for regression coefficients, and standard errors are only off by a factor equal to the ratio of the mean squared error estimate for the reduced model to the original mean squared error estimate for the full model.
执行快速淘汰落后的因素稍微低效的,但数值上是稳定的版本,使用方法的基础上不法分子和阿密特(1978)。此方法使用安装完整的模型,并计算近似Wald统计量的计算条件(限制)的最大似然估计,假设估计的多变量常态。 fastbw删除因素,而不是设计矩阵列。需要多个D.F.将被保留或删除一组。每个变量在函数打印删除统计,并打印后删除变量的模型参数的近似估计。近似是最好的当数被删除的因素并不大。对于ols,近似回归系数的是精确的,并且标准误差只有关闭到原始的平均平方误差预算的完整模型的简化模型的平均平方误差预算的比率相等的一个因素。
If the fit was from ols, fastbw will compute the usual R^2 statistic for each model.
如果是从ols,fastbw将计算通常是R^2统计每个模型的拟合。
用法----------Usage----------
fastbw(fit, rule="aic", type="residual", sls=.05, aics=0, eps=1e-9,
k.aic=2, force=NULL)
## S3 method for class 'fastbw'
print(x, digits=4, ...)
参数----------Arguments----------
参数:fit
fit object with Varcov(fit) defined (e.g., from ols, lrm, cph, psm, glmD)
适合对象Varcov(fit)定义(例如,从ols,lrm,cph,psm,glmD)
参数:rule
Stopping rule. Defaults to "aic" for Akaike's information criterion. Use rule="p" to use P-values
停止规则。默认为"aic" Akaike的信息准则。使用rule="p"使用P值
参数:type
Type of statistic on which to base the stopping rule. Default is "residual" for the pooled residual chi-square. Use type="individual" to use Wald chi-square of individual factors.
统计所依据的停止规则的类型。默认是"residual"的汇集剩余的卡方。使用type="individual"使用Wald卡方个人因素的影响。
参数:sls
Significance level for staying in a model if rule="p". Default is .05.
显着性水平保持在一个模型中,如果rule="p"。默认值是0.05。
参数:aics
For rule="aic", variables are deleted until the chi-square - k.aic times d.f. falls below aics. Default aics is zero to use the ordinary AIC. Set aics to say 10000 to see all variables deleted in order of descending importance.
对于rule="aic",变量被删除,直到卡方 - k.aic倍DF落在下面aics。默认aics是零使用普通AIC的。设置aics说10000,看到所有的变量按降重要性删除。
参数:eps
Singularity criterion, default is 1E-9.
奇异性标准,默认是1E-9。
参数:k.aic
multiplier to compute AIC, default is 2. To use BIC, set k.aic equal to \log(n), where n is the effective sample size (number of events for survival models).
乘数计算AIC,默认值是2。要使用BIC,设置k.aic等于\log(n)的,其中n是有效的样本量(生存模式的事件的数量)。
参数:force
a vector of integers specifying parameters forced to be in the model, not counting intercept(s)
一个向量的整数,指定被迫在模型中的参数,不计拦截(S)
参数:x
result of fastbw
结果fastbw
参数:digits
number of significant digits to print
数量巨大的数字打印
参数:...
ignored
忽视
值----------Value----------
a list with an attribute kept if bw=TRUE, and the following components:
一个属性列表kept如果bw=TRUE,其组成部分如下:
参数:result
matrix of statistics with rows in order of deletion.
为了删除的行矩阵的统计数据。
参数:names.kept
names of factors kept in final model.
因素的名称保留在最终的模型。
参数:factors.kept
the subscripts of factors kept in the final model
保持在最终模型因素的下标
参数:factors.deleted
opposite of factors.kept.
相反的factors.kept。
参数:parms.kept
column numbers in design matrix corresponding to parameters kept in the final model.
保留在最终模型中相应的参数设计矩阵的列数。
参数:parms.deleted
opposite of parms.kept.
相反的parms.kept。
参数:coefficients
vector of approximate coefficients of reduced model.
减少模型的近似系数的矢量。
参数:var
approximate covariance matrix for reduced model.
简化模型的近似协方差矩阵。
参数:Coefficients
matrix of coefficients of all models. Rows correspond to the successive models examined and columns correspond to the coefficients in the full model. For variables not in a particular sub-model (row), the coefficients are zero. </table>
所有型号的系数矩阵。行对应于连续的模型检验和列对应的完整模型中的系数。对于不是在一个特定的子模型(行)的变量,系数均为零。 </ TABLE>
(作者)----------Author(s)----------
Frank Harrell<br>
Department of Biostatistics, Vanderbilt University<br>
f.harrell@vanderbilt.edu
参考文献----------References----------
参见----------See Also----------
rms, ols, lrm, cph, psm, validate, solvet, rmsMisc
rms,ols,lrm,cph,psm,validate,solvet,rmsMisc
实例----------Examples----------
## Not run: [#不运行:]
fastbw(fit, optional.arguments) # print results[打印结果]
z <- fastbw(fit, optional.args) # typically used in simulations[通常模拟中使用的]
lm.fit(X[,z$parms.kept], Y) # least squares fit of reduced model[最小二乘拟合简化模型]
## End(Not run)[#(不执行)]
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。
|