Hi Bill,
I have a feeling the sample is just too constrained and it’s just a really hard matching
problem.
I might recommend trying to do something like exact matching, like CEM (which would
involve potentially coarsening the data, e.g., quintiles of the continuous variables).
Liz
On Apr 16, 2018, at 12:41 PM, William Dudley
<wndudley@uncg.edu<mailto:wndudley@uncg.edu>> wrote:
Matching experts,
I am trying to match a small treatment group of schools (n = 3) with controls using a
ratio of 1:4. I am using Mahal distance.
The co variates are at the school level, but the analyses are planned at the student
level.
The match faces huge challenges to start with.
The three treatment school are among the most extreme in the state with regard to the
covariates:
the highest dropout highest minority, highest poverty and lowest academic performance.
I am getting warnings as follows:
Warning messages:
1: glm.fit: algorithm did not converge
2: glm.fit: fitted probabilities numerically 0 or 1 occurred .
I wonder if there are ways to work around the lack of convergence and fitted probabilities
at the extreme.
In previous work with this group I tried propensity scoring but that was equally
problematic in term sof std differences.
Thanks for any help.
Bill
On Mon, Feb 26, 2018 at 4:31 PM,
<matchit-request@lists.gking.harvard.edu<mailto:matchit-request@lists.gking.harvard.edu>>
wrote:
Send Matchit mailing list submissions to
matchit@lists.gking.harvard.edu<mailto:matchit@lists.gking.harvard.edu
To subscribe or unsubscribe via the World Wide Web, visit
https://lists.gking.harvard.edu/mailman/listinfo/matchit
or, via email, send a message with subject or body 'help' to
matchit-request@lists.gking.harvard.edu<mailto:matchit-request@lists.gking.harvard.edu
You can reach the person managing the list at
matchit-owner@lists.gking.harvard.edu<mailto:matchit-owner@lists.gking.harvard.edu
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Matchit digest..."
Today's Topics:
1. Re: MatchIt Question (Noah Greifer)
----------------------------------------------------------------------
Message: 1
Date: Mon, 26 Feb 2018 21:31:20 +0000
From: Noah Greifer <noah.greifer@gmail.com<mailto:noah.greifer@gmail.com>
To: "Samadi, Mehrdad"
<Mehrdad.Samadi@finra.org<mailto:Mehrdad.Samadi@finra.org>
Cc:
"matchit@lists.gking.harvard.edu<mailto:matchit@lists.gking.harvard.edu>"
<matchit@lists.gking.harvard.edu<mailto:matchit@lists.gking.harvard.edu>
Subject: Re: [matchit] MatchIt Question
Message-ID:
<CAPhhD8ma1szOPw9Zn_-5ber2E7jcovp9YgvK8cngT8bkVkrFxg@mail.gmail.com<mailto:CAPhhD8ma1szOPw9Zn_-5ber2E7jcovp9YgvK8cngT8bkVkrFxg@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
I'm not exactly sure what's going on with your code, but I think you could
run the same analysis without splitting the data set and running 9
separate analyses just be setting exact = "stock" in a single matchit()
call. Then match.data() could be run on the single output object. cobalt's
bal.tab() can help examine balance on the separate groups.
Noah
On Mon, Feb 26, 2018 at 12:20 PM Samadi, Mehrdad
<Mehrdad.Samadi@finra.org<mailto:Mehrdad.Samadi@finra.org>
wrote:
> Hi All,
> I?ve been working with your MatchIT function in R and have a question:
> I?m able to successfully run MatchIT on a list of dataframes using the
> lapply function, which outputs a large list with a number of elements
> corresponding to the number of separate MatchIT procedures that I ran, and
> within each of these elements, 9 sub elements which correspond to the
> standard output of MatchIt.
> While I can successfully extract summary information from this MatchIt
> large list using the summary function and lapply or sapply, I?m having
> difficulty extracting the matrices of matched observations from each
> separate MatchIt procedure. I get the error message, ?invalid subscript
> type ?closure?.
> Some more context is provide below, any insight that you have would be
> helpful. Thanks for your time and consideration.
> The raw dataframe ?sample? has the following structure:
> *stock*
> *firm*
> *var*
> *treated*
> APPL
> A
> 0.5
> 1
> APPL
> B
> 0.54
> 0
> APPL
> C
> 0.33
> 0
> APPL
> D
> 0.38
> 0
> APPL
> E
> 0.88
> 0
> BAC
> A
> 0.5
> 0
> BAC
> B
> 0.54
> 1
> BAC
> C
> 0.33
> 0
> BAC
> D
> 0.38
> 0
> BAC
> E
> 0.88
> 0
> JPM
> A
> 0.5
> 0
> JPM
> B
> 0.54
> 1
> JPM
> C
> 0.33
> 0
> JPM
> D
> 0.38
> 0
> JPM
> E
> 0.88
> 0
> I want to use matchit for each stock separately, so I?m using a
> combination of the split function and lapply:
> X <- split(sample,sample$stock)
> output <- lapply(names(X), function(n) {
> matchit(treated ~
>
var,data=X[[n]],method="nearest",distance="mahalanobis",replace=FALSE,ratio=1)})
> This would output a large list with 3 elements and within each of these 3
> elements, another 9 elements corresponding to MatchIt output.
> I would like to extract the matrices corresponding to the matched
> observations for each stock using match.data by again using lapply or
> sapply, but get the following error:
> sapply(output,match.data)
> Error in X[[n]] : invalid subscript type 'closure'
> Best,
> Mehrdad
> Confidentiality Notice:: This email, including attachments, may include
> non-public, proprietary, confidential or legally privileged information. If
> you are not an intended recipient or an authorized agent of an intended
> recipient, you are hereby notified that any dissemination, distribution or
> copying of the information contained in or transmitted with this e-mail is
> unauthorized and strictly prohibited. If you have received this email in
> error, please notify the sender by replying to this message and permanently
> delete this e-mail, its attachments, and any copies of it immediately. You
> should not retain, copy or use this e-mail or any attachment for any
> purpose, nor disclose all or any part of the contents to any other person.
> Thank you.
> -
> ---
> MatchIt mailing list served by HUIT
> List Address:
matchit@lists.gking.harvard.edu<mailto:matchit@lists.gking.harvard.edu
> Subscribe/Unsubscribe:
http://lists.gking.harvard.edu/mailman/listinfo/ei
> MatchIt Software and Documentation:
http://gking.harvard.edu/matchit/
> Browse/Search <http://gking.harvard.edu/matchit/Browse/Search> List
> Archive:
http://lists.gking.harvard.edu/mailman/private/matchit/
> Matchit mailing list
> Matchit@lists.gking.harvard.edu<mailto:Matchit@lists.gking.harvard.edu
> To unsubscribe from this list or get other information:
>
https://lists.gking.harvard.edu/mailman/listinfo/matchit
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.gking.harvard.edu/pipermail/matchit/attachments/20180226/39dc0f79/attachment.html
------------------------------
_______________________________________________
Matchit mailing list
Matchit@lists.gking.harvard.edu<mailto:Matchit@lists.gking.harvard.edu
To unsubscribe from this list or get other information:
https://lists.gking.harvard.edu/mailman/listinfo/matchit
End of Matchit Digest, Vol 169, Issue 2
***************************************
--
William N. Dudley, PhD
Professor - Public Health Education
The School of Health and Human Sciences
The University of North Carolina at Greensboro
437-L Coleman Building
Greensboro, NC 27402-6170
See my research on
GoogleScholar<https://scholar.google.com/citations?user=ZiYmyb4AAAAJ&…
ResearchGate<https://www.researchgate.net/profile/William_Dudley
VOICE 336.256 2475
[email signature image example.png]
-
---
MatchIt mailing list served by HUIT
List Address:
matchit@lists.gking.harvard.edu<mailto:matchit@lists.gking.harvard.edu
Subscribe/Unsubscribe:
http://lists.gking.harvard.edu/mailman/listinfo/ei
MatchIt Software and Documentation:
http://gking.harvard.edu/matchit/
Browse/Search List Archive:
http://lists.gking.harvard.edu/mailman/private/matchit/
Matchit mailing list
Matchit@lists.gking.harvard.edu<mailto:Matchit@lists.gking.harvard.edu
To unsubscribe from this list or get other information:
https://lists.gking.harvard.edu/mailman/listinfo/matchit