开通博客赚积分
发布资源赚积分
分类

源码开发语言/平台
tuma
文件大小： unknow
源码售价： 5 个金币 积分规则     积分充值
充值1元得10金币   
资源说明：tuma: A graphical user interface for teaching univariate meta-analysis
* tuma: A graphical user interface for teaching univariate meta-analysis 

** Overview
   =tuma= is an add-on package for R  and provides a
   graphical user interface (GUI) to visualize univariate fixed effect (FEM) and
   random effects model (REM) meta-analysis. By interactively manipulating
   effect sizes and standard errors, it is possible to better understand their
   relationship with summary (FEM and REM) and heterogeneity related statistics
   (Q, I^2 and tau^2). Additionally, two plots are created which show a typical
   forest plot as well as a density plot with N(hat{theta}, tau^2). These plots
   also change if the individual effects sizes or standard errors are
   manipulated.  

** Screenshot

   [[file:f_screenshot.png]]

   
** Installation
   The easiest way to install the package is to use the =install_github()=
   function. The =install_github()= function is provided by the =devtools=
   package. 
   
   =tuma= also makes use of the =Bwidget= library. To install the Bwidget
   library under Windows, go to
   http://www.stat.berkeley.edu/users/spector/s133/Bwidget.html and follow the
   instructions. Linux users should be able to install the package =bwidget= via
   their package manager (e.g., under Debain/Ubuntu just use =sudo apt-get
   install bwidget=). 

   If you are on MS Windows, it also seems that you need to have
   Rtools installed. So, you first have to install the package =devtools= (via
   =install.packages("devtools")=, don't get confused by the message "Rtools
   not installed.") and then run:
     1. =library(devtools)= 
     2. =install_github("berndweiss/tuma")=
     3. =library(tuma)=
   
   Binaries for MS Win (.zip) or Linux (.tar.gz) can be found in =/build=

** Usage
   The tuma GUI can be started by typing =tumaGUI()=. 
   
   Unfortunateley, the package does not work well with RStudio! So, do /not/ call the tuma package from within RStudio.

** TODOs
   - fix layout of summary stats
   - add save-plot-button
   - change color of "effect size squares" to gray
   - allow user to plot FEM or REM forest plot; or maybe both... 
   - button to increase ssample size (e.g. double,
   triple sample size)

** Acknowledgements
   This work was supported by a fellowship within the "Postdoc-Programme" of the German Academic Exchange Service (DAAD).

部分文件列表（点击文件名可查看文件内容）
本源码包内暂不包含可直接显示的源代码文件，请下载源码包。