使用metafor包在R中进行元分析

需积分: 10 105 浏览量更新于2024-07-21 1 收藏 800KB PDF 举报

"使用R语言和'metafor'包进行元分析" 元分析是一种统计方法，用于综合和分析来自多个独立研究的结果，以提供一个更强大、更可靠的结论。在R编程环境中，'metafor'包提供了全面的功能，帮助用户进行各种类型的元分析。该包由Rolf R. Wahl于2015年9月28日更新到版本1.9-8，适用于R版本3.2.2及以上。 'metafor'包依赖于R的核心库，如'stats'、'utils'、'graphics'、'grDevices'和'methods'，同时还需要'Matrix'包。它还导入或建议使用其他一系列包，如'lme4'（用于混合效应模型）、'numDeriv'（数值导数计算）、'minqa'（优化算法）等，以支持元分析过程中的各种复杂计算和可视化需求。该包的主要功能包括： 1. 计算各种效应量或结果指标：效应量是衡量研究效果的统计指标，例如风险比、优势比、均值差异等。 2. 固定效应、随机效应和混合效应模型的拟合：这些模型可以帮助处理研究间异质性，随机效应模型允许异质性在研究间存在变异，而混合效应模型则结合了固定效应和随机效应。 3. 中介变量和元回归分析：通过这些方法，可以探究影响研究结果的潜在因素。 4. 创建元分析图表：如森林图、漏斗图、径向图、L'Abbe图和Baujat图，这些图有助于可视化结果和检测发表偏倚。 5. 二项和生存时间数据的特殊方法：包括Mantel-Haenszel方法和Peto's方法，以及混合效应逻辑回归和泊松回归模型，用于处理计数或二元响应数据。 6. 多变量或多级元分析：处理具有多个结局或嵌套数据结构的复杂情况。使用'metafor'包进行元分析时，首先需要安装和加载该包，然后根据研究设计和数据类型选择合适的效应量计算函数。接着，可以使用适当的模型拟合函数（如'revmetan'、'rma.uni'等）进行分析，并通过元回归分析来探索潜在的效应变化。最后，生成图表来展示结果并评估异质性和发表偏倚。 'metafor'包为R用户提供了强大的工具，使得在R中进行复杂的元分析工作变得更为便捷和系统化。无论是初学者还是经验丰富的研究者，都能从中受益，以进行严谨和全面的元分析研究。

16 anova.rma

Value

An object of class "anova.rma". When a single model is speciﬁed, the object is a list containing

the following components:

QM test statistic for the Wald-type test of model coefﬁcients.

QMp p-value for the Wald-type test of model coefﬁcients.

btt indices of the coefﬁcients tested by the Wald-type test.

k number of outcomes included in the model ﬁtting.

p number of coefﬁcients in the model (including the intercept).

m number of coefﬁcients included in the Wald-type test of coefﬁcients.

... some additional elements/values.

When two models are speciﬁed, the object is a list containing the following components:

fit.stats.f log-likelihood, deviance, AIC, BIC, and AICc for the full model.

fit.stats.r log-likelihood, deviance, AIC, BIC, and AICc for the reduced model.

p.f number of parameters in the full model.

p.r number of parameters in the reduced model.

LRT likelihood ratio test statistic.

pval p-value for the likelihood ratio test.

QE.f test statistic for the test of (residual) heterogeneity from the full model.

QE.r test statistic for the test of (residual) heterogeneity from the reduced model.

tau2.f estimated τ

value from the full model. NA for "rma.mv" objects.

tau2.r estimated τ

value from the reduced model. NA for "rma.mv" objects.

R2 amount of (residual) heterogeneity in the reduced model that is accounted for

in the full model (in percent). NA for ﬁxed-effects models, if the amount of

heterogeneity in the reduced model is equal to zero, or for "rma.mv" objects.

This can be regarded as a pseudo R

statistic (Raudenbush, 2009). Note that the

value may not be very accurate unless k is large (Lopez-Lopez et al., 2014).

... some additional elements/values.

The results are formated and printed with the print.anova.rma function.

Note

In principle, one can consider likelihood ratio tests for the amount of (residual) heterogeneity in

random- and mixed-effects models. The full model should then be ﬁtted with either method="ML"

or method="REML" and the reduced model with method="FE". The p-value from that test is based

on a chi-square distribution with 1 degree of freedom, but actually needs to be adjusted for the fact

that the parameter (i.e., τ

) falls on the boundary of the parameter space under the null hypothesis.

Moreover, the Q-test usually keeps better control of the Type I error rate and therefore should be

preferred (see Viechtbauer, 2007, for more details).

baujat 19

in the model ﬁtting) on the y-axis. The same type of plot can be produced by ﬁrst ﬁtting a ﬁxed-

effects model with either the rma.uni (using method="FE"), rma.mh, or rma.peto functions and

then passing the ﬁtted model object to the baujat function.

For models ﬁtted with the rma.uni function (which may involve moderators and/or may be random/mixed-

effects models), the idea underlying this type of plot can be generalized as follows: The x-axis then

corresponds to the squared Pearson residual of a study, while the y-axis corresponds to the stan-

dardized squared difference between the predicted/ﬁtted value for the study with and without the

study included in the model ﬁtting. Therefore, for a ﬁxed-effect with moderators model, the x-axis

corresponds to the contribution of the study to the QE-test statistic for residual heterogeneity.

The points are labeled according to the slab argument (see model ﬁtting functions). If slab was

unspeciﬁed, then the numbers 1 through k are used by default.

Value

A data frame with components:

x the x coordinates of the points that were plotted.

y the y coordinates of the points that were plotted.

slab study labels of the points that were plotted.

Note that the data frame is returned invisibly.

Author(s)

Wolfgang Viechtbauer <wvb@metafor-project.org>

package website: http://www.metafor-project.org/

author homepage: http://www.wvbauer.com/

References

Baujat, B., Mahe, C., Pignon, J.-P., & Hill, C. (2002). A graphical method for exploring hetero-

geneity in meta-analyses: Application to a meta-analysis of 65 trials. Statistics in Medicine, 21(18),

2641–2652.

Viechtbauer, W. (2010). Conducting meta-analyses in R with the metafor package. Journal of

Statistical Software, 36(3), 1–48. http://www.jstatsoft.org/v36/i03/.

使用metafor包在R中进行元分析

Introductory.Fisheries.Analyses.with.R

NIST SP800-14.pdf

Directive Electromagnetic Scattering by an Infinite Conducting Cylinder Coated with Left-handed Material

Sinatkas et al. - 2017 - Transparent conducting oxide

NIST Standard(800-30)：Guide for Conducting Risk Assessments of Information Security Risk - 完整英文电子版（95页）.pdf

Scattering of a Plane Wave by an Anisotropic Plasma-Coated Conducting Sphere

ASTM E691-20 Standard Practice for Conducting an lnterlaboratory

ASTM E1169-21 Standard Practice for Conducting Ruggedness Tests

NIST Standard(800-30)：Guide for Conducting Risk Assessments of I

High-Order Small Perturbation Method of Scattering from Conducting Random Rough Surfaces: TM Case

最新资源