Difference: MatchedRandomization (1 vs. 30)

Revision 30
17 Dec 2015 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

Line: 45 to 45
 

Known Bugs and Desired Features

Note many of the bugs and desired features have been fixed and added. We'll cull the following list down in the future.
Added:
>
>
  1. Create bib so cite('nbpMatching') works. In the meantime, the standard format is encouraged, e.g.
    Beck C, Lu B, Greevy R. (2015). nbpMatching: Functions for Optimal Non-Bipartite Matching. R package version 1.4.5. https://cran.r-project.org/web/packages/nbpMatching
 
  1. Documentation for nbpMatching needs to be updated and should point to this page and include the PDS paper reference. Tutorial examples on this page should be expanded.
  2. The qom() function calculates the quantiles of the absolute mean differences for two treatment arms over the randomization space. It offers the option of including or excluding subjects matched to phantoms. The calculations when including subjects matched to phantoms are slightly off.
  3. Modify qom() to calculate the AMD_100 by placing the max of each pair in group A and the min of each pair in group B, and calculating the AMD on that worst case; as opposed to taking the max of the sampled randomization set which will likely miss that worst case.
Revision 29
29 Jan 2015 - Main.ColeBeck
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

Line: 7 to 7
  The current version of the R package nbpMatching is currently available at:
Changed:
<
<
  • GitHub TBD.
>
>
  Active development occurs on R-Forge and GitHub with stable versions being pushed to cran periodically. Installing a current version of nbpMatching may require updating R.

Line: 18 to 18
  install.packages("nbpMatching", repos="http://R-Forge.R-project.org")

# installing from GitHub
Changed:
<
<
## TBD
>
>
library(devtools) install_github('couthcommander/nbpMatching')
 

# check your version library( nbpMatching )
Revision 28
29 Jan 2015 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

Changed:
<
<

Matched Randomization R Package

The current version (1.4.0) of the R package nbpMatching is currently available at CRAN. We develop the package on R-Forge. Installing the current version may require updating R.
>
>

Matched Randomization R Package nbpMatching

The current version of the R package nbpMatching is currently available at: Active development occurs on R-Forge and GitHub with stable versions being pushed to cran periodically. Installing a current version of nbpMatching may require updating R.
 
Changed:
<
<
# install the latest stable version from CRAN
>
>
# installing from CRAN
  install.packages( "nbpMatching" )
Changed:
<
<
# install the latest beta version from R-Forge
>
>
# installing from R-Forge
  install.packages("nbpMatching", repos="http://R-Forge.R-project.org")
Added:
>
>
# installing from GitHub ## TBD
  # check your version library( nbpMatching ) library( help='nbpMatching' )
Line: 35 to 39
 
Deleted:
<
<
 

What's New (version history)

Changed:
<
<
  • 1.4.0 -- Multiple performance improvements.
  • 1.3.5 -- Corrected a bug that would cause fill.missing() to fail the first time it was run.
  • 1.3.4 -- Much faster gendistance function.
  • 1.3.3 -- Works with R 2.14 and fixed possible crash when using prevent option with few variables.
  • More details and source code may be viewed at R-Forge.
>
>
  • The latest improvements and source code may be viewed at R-Forge.
 

Known Bugs and Desired Features

Added:
>
>
Note many of the bugs and desired features have been fixed and added. We'll cull the following list down in the future.
 
  1. Documentation for nbpMatching needs to be updated and should point to this page and include the PDS paper reference. Tutorial examples on this page should be expanded.
  2. The qom() function calculates the quantiles of the absolute mean differences for two treatment arms over the randomization space. It offers the option of including or excluding subjects matched to phantoms. The calculations when including subjects matched to phantoms are slightly off.
  3. Modify qom() to calculate the AMD_100 by placing the max of each pair in group A and the min of each pair in group B, and calculating the AMD on that worst case; as opposed to taking the max of the sampled randomization set which will likely miss that worst case.
Revision 27
22 Jan 2014 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

Line: 31 to 31
 
Added:
>
>
What is that "Note: Distances scaled" warning all about?
 

What's New (version history)

Revision 26
11 Sep 2013 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

Matched Randomization R Package

Changed:
<
<
The current version (1.3.5) of the updated R package nbpMatching is currently available at R-Forge and will be available on CRAN soon. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
>
>
The current version (1.4.0) of the R package nbpMatching is currently available at CRAN. We develop the package on R-Forge. Installing the current version may require updating R.
 
Changed:
<
<
# install from R-Forge (working on 2012-05-29. Requires R version 2.15 or later, # and requires installing Hmisc from CRAN separately.) install.packages("nbpMatching", repos="http://R-Forge.R-project.org") # install from CRAN (currently two versions behind, # may have trouble with the prevent option is some settings)
>
>
# install the latest stable version from CRAN
  install.packages( "nbpMatching" )
Changed:
<
<
# load package once it's installed
>
>

# install the latest beta version from R-Forge install.packages("nbpMatching", repos="http://R-Forge.R-project.org")

# check your version
  library(nbpMatching)
Changed:
<
<
# open up documentation; version number at the bottom of the page ?nbpMatching
>
>
library( help='nbpMatching' )
 

nbpMatching Tutorials

Line: 35 to 34
 

What's New (version history)

Added:
>
>
  • 1.4.0 -- Multiple performance improvements.
 
  • 1.3.5 -- Corrected a bug that would cause fill.missing() to fail the first time it was run.
  • 1.3.4 -- Much faster gendistance function.
  • 1.3.3 -- Works with R 2.14 and fixed possible crash when using prevent option with few variables.
Revision 25
06 Sep 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

Line: 25 to 27
 
Added:
>
>

This tutorial shows how to use the fill.missing() function to impute missing values and create missingness indicator variables.

 

What's New (version history)

  • 1.3.5 -- Corrected a bug that would cause fill.missing() to fail the first time it was run.
  • 1.3.4 -- Much faster gendistance function.
Revision 24
18 Jul 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

Line: 34 to 34
 

Known Bugs and Desired Features

  1. Documentation for nbpMatching needs to be updated and should point to this page and include the PDS paper reference. Tutorial examples on this page should be expanded.
  2. The qom() function calculates the quantiles of the absolute mean differences for two treatment arms over the randomization space. It offers the option of including or excluding subjects matched to phantoms. The calculations when including subjects matched to phantoms are slightly off.
Added:
>
>
  1. Modify qom() to calculate the AMD_100 by placing the max of each pair in group A and the min of each pair in group B, and calculating the AMD on that worst case; as opposed to taking the max of the sampled randomization set which will likely miss that worst case.
 
  1. It would be nice for the the quality of matches function, qom, to return the standard deviation for the AMDs and to take the number of simulations to run as an input, e.g. someone could input 100,000 sims if 10,000 sims wasn't enough for the needed level of precision.
  2. Consider adding the SEs for the AMDs to the automatically generated benchmarking balance tables. This could get cluttered, but maybe a max SE for each variable could be included as a footnote, excluding the SEs for AMD_0 and AMD_100. The command hdquantile in Hmisc is pretty fast and will give the quantile with the se.
  3. Allow individual weight specification of missingness indicators. If length(missing.weight)==1, use the same weight for all missingness indicators. If length(missing.weight) == the number of missingness indicators, use the respective weights. Else, warn the user that "the number of elements in missing.weight does not equal the number of variables with missingness".
Revision 23
13 Jul 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

Line: 33 to 33
 

Known Bugs and Desired Features

  1. Documentation for nbpMatching needs to be updated and should point to this page and include the PDS paper reference. Tutorial examples on this page should be expanded.
Added:
>
>
  1. The qom() function calculates the quantiles of the absolute mean differences for two treatment arms over the randomization space. It offers the option of including or excluding subjects matched to phantoms. The calculations when including subjects matched to phantoms are slightly off.
 
  1. It would be nice for the the quality of matches function, qom, to return the standard deviation for the AMDs and to take the number of simulations to run as an input, e.g. someone could input 100,000 sims if 10,000 sims wasn't enough for the needed level of precision.
Deleted:
<
<
  1. Allow individual weight specification of missingness indicators. If length(missing.weight)==1, use the same weight for all missingness indicators. If length(missing.weight) == the number of missingness indicators, use the respective weights. Else, warn the user that "the number of elements in missing.weight does not equal the number of variables with missingness".
 
  1. Consider adding the SEs for the AMDs to the automatically generated benchmarking balance tables. This could get cluttered, but maybe a max SE for each variable could be included as a footnote, excluding the SEs for AMD_0 and AMD_100. The command hdquantile in Hmisc is pretty fast and will give the quantile with the se.
Added:
>
>
  1. Allow individual weight specification of missingness indicators. If length(missing.weight)==1, use the same weight for all missingness indicators. If length(missing.weight) == the number of missingness indicators, use the respective weights. Else, warn the user that "the number of elements in missing.weight does not equal the number of variables with missingness".
  2. We want to revisit how smoothly the package and webapp handle perfectly collinear variables. This is most likely to occur in the generated missingness indicators, e.g. systolic.missing and diastolic.missing are likely to be perfectly collinear.
  3. Look at how fill.missing() handles the id column. Allow fill.missing() to take idcol=# as an option so it can be more easily used independently from gendistance().
 



Revision 22
29 Jun 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

Line: 29 to 29
 
  • 1.3.5 -- Corrected a bug that would cause fill.missing() to fail the first time it was run.
  • 1.3.4 -- Much faster gendistance function.
  • 1.3.3 -- Works with R 2.14 and fixed possible crash when using prevent option with few variables.
Added:
>
>
  • More details and source code may be viewed at R-Forge.
 

Known Bugs and Desired Features

  1. Documentation for nbpMatching needs to be updated and should point to this page and include the PDS paper reference. Tutorial examples on this page should be expanded.
Revision 21
29 Jun 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

Matched Randomization R Package

Changed:
<
<
The current version (1.3.4) of the updated R package nbpMatching is currently available at R-Forge and will be available on CRAN soon. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
>
>
The current version (1.3.5) of the updated R package nbpMatching is currently available at R-Forge and will be available on CRAN soon. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
  # install from R-Forge (working on 2012-05-29. Requires R version 2.15 or later, # and requires installing Hmisc from CRAN separately.)
Line: 25 to 25
 
Changed:
<
<

Version History

>
>

What's New (version history)

  • 1.3.5 -- Corrected a bug that would cause fill.missing() to fail the first time it was run.
 
  • 1.3.4 -- Much faster gendistance function.
  • 1.3.3 -- Works with R 2.14 and fixed possible crash when using prevent option with few variables.
Revision 20
29 May 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

Line: 6 to 6
 

Matched Randomization R Package

The current version (1.3.4) of the updated R package nbpMatching is currently available at R-Forge and will be available on CRAN soon. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
Changed:
<
<
# install from R-Forge (working on 2012-05-29. Requires R > 2.15, and requires installing Hmisc from CRAN separately.)
>
>
# install from R-Forge (working on 2012-05-29. Requires R version 2.15 or later, # and requires installing Hmisc from CRAN separately.)
  install.packages("nbpMatching", repos="http://R-Forge.R-project.org")
Changed:
<
<
# install from CRAN (currently two versions behind, may have trouble with the prevent option is some settings)
>
>
# install from CRAN (currently two versions behind, # may have trouble with the prevent option is some settings)
  install.packages( "nbpMatching" ) # load package once it's installed library(nbpMatching)
Revision 19
29 May 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

Matched Randomization R Package

Changed:
<
<
The current version (1.3.3) of the updated R package nbpMatching is currently available at R-Forge and will be available on CRAN soon. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
>
>
The current version (1.3.4) of the updated R package nbpMatching is currently available at R-Forge and will be available on CRAN soon. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
 
Changed:
<
<
# install from R-Forge (working on 2012-03-23, may need to install Hmisc from CRAN separately)
>
>
# install from R-Forge (working on 2012-05-29. Requires R > 2.15, and requires installing Hmisc from CRAN separately.)
  install.packages("nbpMatching", repos="http://R-Forge.R-project.org")
Changed:
<
<
# install from CRAN (currently one version behind, may have trouble with the prevent option is some settings)
>
>
# install from CRAN (currently two versions behind, may have trouble with the prevent option is some settings)
  install.packages( "nbpMatching" ) # load package once it's installed library(nbpMatching)
Line: 23 to 23
 
Added:
>
>

Version History

  • 1.3.4 -- Much faster gendistance function.
  • 1.3.3 -- Works with R 2.14 and fixed possible crash when using prevent option with few variables.
 

Known Bugs and Desired Features

  1. Documentation for nbpMatching needs to be updated and should point to this page and include the PDS paper reference. Tutorial examples on this page should be expanded.
  2. It would be nice for the the quality of matches function, qom, to return the standard deviation for the AMDs and to take the number of simulations to run as an input, e.g. someone could input 100,000 sims if 10,000 sims wasn't enough for the needed level of precision.
Revision 18
15 May 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"
Added:
>
>

Reweighted Mahalanobis Distance Matching in Observational Studies and Randomized Trials

 

Matched Randomization R Package

The current version (1.3.3) of the updated R package nbpMatching is currently available at R-Forge and will be available on CRAN soon. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
Line: 37 to 40
  The web application provides links to download the generated distance matrix, a full and a reduced table of the optimal matches, to assess the quality of the matching if being used for a randomized trial, and to perform the randomization within pairs. When randomizing, the application assigns treatments "A" and "B" and allows the user to specify a randomization seed for reproducibility.



Changed:
<
<

Tutorial

>
>

WebApp Tutorial

 

Uploading Your Covariate Matrix

Unless you have created your own distance matrix already, you begin by uploading a dataset with your covariates. They need to be in a comma separated file, e.g. ClusterRandomizedExample2.csv. With the exception of the variable names and ID column values, all variable values should be numeric. Any non-numeric value including "NA", ".", and a blank "" will be treated as missing values. Categorical variables should be broken into indicators. For example, a location variable with four levels (Northern, Southern, Western, MidWestern) should be made into three indicator variables (Southern, Western, MidWestern; Northern is referent).
Revision 17
23 Mar 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization R Package

Changed:
<
<
The current version (1.3.2) of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
>
>
The current version (1.3.3) of the updated R package nbpMatching is currently available at R-Forge and will be available on CRAN soon. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
 
Changed:
<
<
# install from R-Forge (working on 2011-11-28, may need to install Hmisc from CRAN separately)
>
>
# install from R-Forge (working on 2012-03-23, may need to install Hmisc from CRAN separately)
  install.packages("nbpMatching", repos="http://R-Forge.R-project.org")
Changed:
<
<
# install from CRAN (working on 2011-11-28, may need to change mirror site to install Hmisc)
>
>
# install from CRAN (currently one version behind, may have trouble with the prevent option is some settings)
  install.packages( "nbpMatching" ) # load package once it's installed library(nbpMatching)
Revision 16
25 Feb 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization R Package

The current version (1.3.2) of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
Line: 24 to 24
 
  1. Documentation for nbpMatching needs to be updated and should point to this page and include the PDS paper reference. Tutorial examples on this page should be expanded.
  2. It would be nice for the the quality of matches function, qom, to return the standard deviation for the AMDs and to take the number of simulations to run as an input, e.g. someone could input 100,000 sims if 10,000 sims wasn't enough for the needed level of precision.
  3. Allow individual weight specification of missingness indicators. If length(missing.weight)==1, use the same weight for all missingness indicators. If length(missing.weight) == the number of missingness indicators, use the respective weights. Else, warn the user that "the number of elements in missing.weight does not equal the number of variables with missingness".
Changed:
<
<
  1. Consider adding the SEs for the AMDs to the automatically generated benchmarking balance tables. This could get cluttered, but maybe a max SE for each variable could be included as a footnote, excluding the SEs for AMD_0 and AMD_100.
>
>
  1. Consider adding the SEs for the AMDs to the automatically generated benchmarking balance tables. This could get cluttered, but maybe a max SE for each variable could be included as a footnote, excluding the SEs for AMD_0 and AMD_100. The command hdquantile in Hmisc is pretty fast and will give the quantile with the se.
 



Revision 15
18 Feb 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization R Package

The current version (1.3.2) of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
Line: 21 to 21
 

Known Bugs and Desired Features

Added:
>
>
  1. Documentation for nbpMatching needs to be updated and should point to this page and include the PDS paper reference. Tutorial examples on this page should be expanded.
 
  1. It would be nice for the the quality of matches function, qom, to return the standard deviation for the AMDs and to take the number of simulations to run as an input, e.g. someone could input 100,000 sims if 10,000 sims wasn't enough for the needed level of precision.
Added:
>
>
  1. Allow individual weight specification of missingness indicators. If length(missing.weight)==1, use the same weight for all missingness indicators. If length(missing.weight) == the number of missingness indicators, use the respective weights. Else, warn the user that "the number of elements in missing.weight does not equal the number of variables with missingness".
  2. Consider adding the SEs for the AMDs to the automatically generated benchmarking balance tables. This could get cluttered, but maybe a max SE for each variable could be included as a footnote, excluding the SEs for AMD_0 and AMD_100.
 



Revision 14
17 Jan 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization R Package

The current version (1.3.2) of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
Line: 17 to 17
  The first tutorial focuses on using nbpMatching for an observational study. We recommend doing this tutorial first even if doing a randomized trial. It teaches several important features of nbpMatching package. This tutorial shows how to use nbpMatching to create matched triplets for a randomized trial.
Changed:
<
<
>
>
 

Known Bugs and Desired Features

  1. It would be nice for the the quality of matches function, qom, to return the standard deviation for the AMDs and to take the number of simulations to run as an input, e.g. someone could input 100,000 sims if 10,000 sims wasn't enough for the needed level of precision.
Revision 13
17 Jan 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization R Package

The current version (1.3.2) of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
Line: 19 to 19
  This tutorial shows how to use nbpMatching to create matched triplets for a randomized trial.
Added:
>
>

Known Bugs and Desired Features

  1. It would be nice for the the quality of matches function, qom, to return the standard deviation for the AMDs and to take the number of simulations to run as an input, e.g. someone could input 100,000 sims if 10,000 sims wasn't enough for the needed level of precision.
 



Revision 12
17 Jan 2012 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization R Package

The current version (1.3.2) of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
Line: 20 to 20
 
Changed:
<
<
>
>





 

Matched Randomization Web Application

The web application http://data.vanderbilt.edu/rapache/megamatch allows users to upload a dataset of covariates on which to match (in csv format) and creates the set of optimally matched pairs that minimizes the average reweighted Mahalanobis distance between pairs. Users may choose the weights for each covariate, may select covariates to be transformed to ranks, may prevent certain matches from forming, and may select a number of units to optimally discard. If the dataset contains missing values, users may control whether to match on imputed values, match on missingness patterns, or a weighted combination of the two. Optionally, users may directly upload a distance matrix on which to match.
Revision 11
30 Nov 2011 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization R Package

The current version (1.3.2) of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
Line: 14 to 14
 

nbpMatching Tutorials

Changed:
<
<
The first tutorial focuses on using nbpMatching for an observational study. We recommend doing this tutorial first even if doing a randomized trial.
>
>
The first tutorial focuses on using nbpMatching for an observational study. We recommend doing this tutorial first even if doing a randomized trial. It teaches several important features of nbpMatching package.
  This tutorial shows how to use nbpMatching to create matched triplets for a randomized trial.
Revision 10
28 Nov 2011 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization R Package

Changed:
<
<
The current version of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
>
>
The current version (1.3.2) of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
 
Changed:
<
<
# install from R-Forge; this has been more reliable
>
>
# install from R-Forge (working on 2011-11-28, may need to install Hmisc from CRAN separately)
  install.packages("nbpMatching", repos="http://R-Forge.R-project.org")
Changed:
<
<
# install from CRAN; this has been less reliable
>
>
# install from CRAN (working on 2011-11-28, may need to change mirror site to install Hmisc)
  install.packages( "nbpMatching" ) # load package once it's installed library(nbpMatching)
Changed:
<
<
# open up documentation; includes version number
>
>
# open up documentation; version number at the bottom of the page
  ?nbpMatching
Revision 9
28 Nov 2011 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization R Package

The current version of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.
Line: 17 to 17
  The first tutorial focuses on using nbpMatching for an observational study. We recommend doing this tutorial first even if doing a randomized trial. This tutorial shows how to use nbpMatching to create matched triplets for a randomized trial.
Changed:
<
<
>
>
 

Revision 8
16 Nov 2011 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization R Package

Changed:
<
<
The current version of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page.
>
>
The current version of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page. The following commands install the package and check which version you have installed.

<-- 
 -->
# install from R-Forge; this has been more reliable
install.packages("nbpMatching", repos="http://R-Forge.R-project.org")
# install from CRAN; this has been less reliable
install.packages( "nbpMatching" ) 
# load package once it's installed
library(nbpMatching)
# open up documentation; includes version number
?nbpMatching
<-- 
-->

nbpMatching Tutorials

The first tutorial focuses on using nbpMatching for an observational study. We recommend doing this tutorial first even if doing a randomized trial. This tutorial shows how to use nbpMatching to create matched triplets for a randomized trial.
 

Matched Randomization Web Application

The web application http://data.vanderbilt.edu/rapache/megamatch allows users to upload a dataset of covariates on which to match (in csv format) and creates the set of optimally matched pairs that minimizes the average reweighted Mahalanobis distance between pairs. Users may choose the weights for each covariate, may select covariates to be transformed to ranks, may prevent certain matches from forming, and may select a number of units to optimally discard. If the dataset contains missing values, users may control whether to match on imputed values, match on missingness patterns, or a weighted combination of the two. Optionally, users may directly upload a distance matrix on which to match.
Line: 47 to 65
 

Deleted:
<
<

Matching in an Observational Study

MatchingInAnObservationalStudyDemo
 

Revision 7
20 Oct 2011 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization R Package

The current version of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page.
Line: 47 to 47
 

Added:
>
>

Matching in an Observational Study

MatchingInAnObservationalStudyDemo

 

META FILEATTACHMENT attachment="ClusterRandomizedExample2.csv" attr="h" comment="" date="1307388406" name="ClusterRandomizedExample2.csv" path="ClusterRandomizedExample2.csv" size="1512" stream="IO::File=GLOB(0x95946bc)" tmpFilename="/tmp/8iJz28eNpI" user="RobertGreevy" version="1"
Revision 6
10 Oct 2011 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization R Package

Changed:
<
<
The beta version of the updated R package nbpMatching is currently available at R-Forge. The official update will be made available on CRAN at a future date. R-Forge automatically generates binaries for Windows and MacOS and may be delayed up to 1-2 days after a recent update the Linux source.
>
>
The current version of the updated R package nbpMatching is currently available at R-Forge and on CRAN. We will be improving the documentation in the next update and on this page.
 

Matched Randomization Web Application

The web application http://data.vanderbilt.edu/rapache/megamatch allows users to upload a dataset of covariates on which to match (in csv format) and creates the set of optimally matched pairs that minimizes the average reweighted Mahalanobis distance between pairs. Users may choose the weights for each covariate, may select covariates to be transformed to ranks, may prevent certain matches from forming, and may select a number of units to optimally discard. If the dataset contains missing values, users may control whether to match on imputed values, match on missingness patterns, or a weighted combination of the two. Optionally, users may directly upload a distance matrix on which to match.
Revision 5
31 Aug 2011 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"
Added:
>
>

Matched Randomization R Package

The beta version of the updated R package nbpMatching is currently available at R-Forge. The official update will be made available on CRAN at a future date. R-Forge automatically generates binaries for Windows and MacOS and may be delayed up to 1-2 days after a recent update the Linux source.
 

Matched Randomization Web Application

The web application http://data.vanderbilt.edu/rapache/megamatch allows users to upload a dataset of covariates on which to match (in csv format) and creates the set of optimally matched pairs that minimizes the average reweighted Mahalanobis distance between pairs. Users may choose the weights for each covariate, may select covariates to be transformed to ranks, may prevent certain matches from forming, and may select a number of units to optimally discard. If the dataset contains missing values, users may control whether to match on imputed values, match on missingness patterns, or a weighted combination of the two. Optionally, users may directly upload a distance matrix on which to match.
Revision 4
07 Jun 2011 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization Web Application

The web application http://data.vanderbilt.edu/rapache/megamatch allows users to upload a dataset of covariates on which to match (in csv format) and creates the set of optimally matched pairs that minimizes the average reweighted Mahalanobis distance between pairs. Users may choose the weights for each covariate, may select covariates to be transformed to ranks, may prevent certain matches from forming, and may select a number of units to optimally discard. If the dataset contains missing values, users may control whether to match on imputed values, match on missingness patterns, or a weighted combination of the two. Optionally, users may directly upload a distance matrix on which to match.
Line: 22 to 22
 
8 62.6 24.5 100.8 7.3 8.9 53.8 28.2 2.2 649
9 62.7 28.2 98.5 7.8 10.3 52.5 27.3 13.3 2241
10 62.9 28.2 107.3 7.3 9 64.4 40.4 17.1 3066
Changed:
<
<
...
>
>
...
 

Create Your Distance Matrix

Changed:
<
<
If you have an ID column, click the appropriate radio button to identify that column. Check any columns you wish to be transformed to ranked values before matching. Check any columns you wish to use to prevent matches.
>
>
If you have an ID column, click the appropriate radio button to identify that column. Check any columns you wish to be transformed to ranked values before matching. Check any columns indicating matches to prevent. Set desired weights for variables and missingness; weights may be any number greater than or equal to 0. Set the number of units you wish to drop. These units will be matched to "phantoms" as indicated in the results.

Create Your Matches

Using your distance matrix, create your matched pairs via optimal nonbipartite matching.

Examine The Quality of Your Matches

This will show the upper percentiles for the absolute mean differences for your covariates over 10,000 possible randomizations.

Randomize

Create the official randomization, setting the seed or making note of the default seed for reproducibility.
 

Revision 3
06 Jun 2011 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization Web Application

The web application http://data.vanderbilt.edu/rapache/megamatch allows users to upload a dataset of covariates on which to match (in csv format) and creates the set of optimally matched pairs that minimizes the average reweighted Mahalanobis distance between pairs. Users may choose the weights for each covariate, may select covariates to be transformed to ranks, may prevent certain matches from forming, and may select a number of units to optimally discard. If the dataset contains missing values, users may control whether to match on imputed values, match on missingness patterns, or a weighted combination of the two. Optionally, users may directly upload a distance matrix on which to match.
Line: 25 to 25
  ...
Changed:
<
<
>
>

Create Your Distance Matrix

If you have an ID column, click the appropriate radio button to identify that column. Check any columns you wish to be transformed to ranked values before matching. Check any columns you wish to use to prevent matches.
 

Revision 2
06 Jun 2011 - Main.RobertGreevy
Line: 1 to 1
 
META TOPICPARENT name="RobertGreevy"

Matched Randomization Web Application

The web application http://data.vanderbilt.edu/rapache/megamatch allows users to upload a dataset of covariates on which to match (in csv format) and creates the set of optimally matched pairs that minimizes the average reweighted Mahalanobis distance between pairs. Users may choose the weights for each covariate, may select covariates to be transformed to ranks, may prevent certain matches from forming, and may select a number of units to optimally discard. If the dataset contains missing values, users may control whether to match on imputed values, match on missingness patterns, or a weighted combination of the two. Optionally, users may directly upload a distance matrix on which to match.

The web application provides links to download the generated distance matrix, a full and a reduced table of the optimal matches, to assess the quality of the matching if being used for a randomized trial, and to perform the randomization within pairs. When randomizing, the application assigns treatments "A" and "B" and allows the user to specify a randomization seed for reproducibility.
Added:
>
>


Tutorial

Uploading Your Covariate Matrix

Unless you have created your own distance matrix already, you begin by uploading a dataset with your covariates. They need to be in a comma separated file, e.g. ClusterRandomizedExample2.csv. With the exception of the variable names and ID column values, all variable values should be numeric. Any non-numeric value including "NA", ".", and a blank "" will be treated as missing values. Categorical variables should be broken into indicators. For example, a location variable with four levels (Northern, Southern, Western, MidWestern) should be made into three indicator variables (Southern, Western, MidWestern; Northern is referent).
 
Added:
>
>
The example dataset ClusterRandomizedExample2.csv can be opened with any statistical software, spreadsheet program, or text editor. The first ten rows look like the following.
SiteStudyID AgeMean PercentPosCVDHistory LDLMean A1cMean A1c90thPercentile PercentOnStatins PercentSulfonylureaUsers PercentAfricanAmerican NumberOfPatients
1 58.3 20.8 117.5 7.6 9.9 58.5 35.8 34.6 1421
2 NA 19.8 114.3 7.5 9.7 47.7 25.1 13.7 1975
3 61.1 18.1 106.3 7.3 9.2 47.2 23.9 4.6 1371
4 61.6 27.6 109 7.6 10 52.4 44.5 28.7 1793
5 61.8 22.2 103.4 7.7 10.3 54.7 44.7 10.4 2218
6 62.3 34.6 NA 7.4 9.2 58.8 40.9 2 1793
7 62.5 27.1 115.5 6.9 8.3 56.5 31 21.6 667
8 62.6 24.5 100.8 7.3 8.9 53.8 28.2 2.2 649
9 62.7 28.2 98.5 7.8 10.3 52.5 27.3 13.3 2241
10 62.9 28.2 107.3 7.3 9 64.4 40.4 17.1 3066
...
 

Added:
>
>

META FILEATTACHMENT attachment="ClusterRandomizedExample2.csv" attr="h" comment="" date="1307388406" name="ClusterRandomizedExample2.csv" path="ClusterRandomizedExample2.csv" size="1512" stream="IO::File=GLOB(0x95946bc)" tmpFilename="/tmp/8iJz28eNpI" user="RobertGreevy" version="1"
 
This site is powered by FoswikiCopyright © 2013 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback