Perhaps I am misunderstanding but in tidyverse dialect I think this works:
library(tidyverse)
tibble(subject1 = 1:3,
date = as.Date(c("01/01/2020", "01/01/2021", NA_Date_),
format = "%d/%m/%Y")) %>%
mutate(rowN1 = row_number()) -> tmpTib1
tibble(subject2 = c(100, 100, 101),
date = as.Date(c("01/01/2020", "01/01/2021", "02/01/2021"),
format = "%d/%m/%Y")) %>%
mutate(rowN2 = row_number()) -> tmpTib2
tmpTib1
### gives
# A tibble: 3 × 3
# subject1 date rowN1
# <int> <date> <int>
#1 1 2020-01-01 1
#2 2 2021-01-01 2
#3 3 NA 3
tmpTib2
### gives
# A tibble: 3 × 3
# subject2 date rowN2
# <dbl> <date> <int>
#1 100 2020-01-01 1
#2 100 2021-01-01 2
#3 101 2021-01-02 3
tmpTib2 %>%
left_join(tmpTib1, by = "date") %>%
group_by(subject2) %>%
ungroup() %>%
filter(row_number() == 1)
### gives
# A tibble: 1 × 5
# subject2 date rowN2 subject1 rowN1
# <dbl> <date> <int> <int> <int>
#1 100 2020-01-01 1 1 1
I think this would be highly inefficient if file 2 is huge with many records
with the same date (matching a date from file1).
I have also assumed that you just want the first match in the default file sort
order but you could rearrange the files to get the
order you want before doing the above.
My rowN variables are not needed!
But I suspect I am misunderstanding and I am sure there are other ways to do
this without tidyverse.
Very best all,
Chris
On 22/07/2026 20:58, Sorkin, John wrote:
I am not showing code, because I don't know how to write the code. I know that
MatchIt performs propensity score matching, but I don't know how to use the
function to perform the match I want to conduct.
I have two files
File 1: a list of n records, 1 record/subject. Each record has a subject
number and a date
File 2: a list of m records m>>n with multiple records/subject but each record
has a distinct date. All records for a given subject in file 2 have the same subject
number, but each record has a different date.
I want to match each record from file 1 with a single record from file 2 based
on date. Once a record from file 2 is matched with a record from file 1, I want
to remove all records from file 2 for the subject who was used in the match
with the subject from file 1 so that the subject from file 2 will not be used
in any future match.
Example
file 1 file 2
subject date subject date
1 01/01/2020 100 01/01/2020
2 01/01/2021 100 01/10/2021
3 101 02/01/2021
Subject 1 from file 1 is a perfect match for subject 100 from file2. The subjects are matched.
All data from file 2 is removed for subject 100.
Subject 2 from file 1 is then matched with subject 101 from file 2.
If all records from subject 100 had not been removed from file2, subject 2
would have been matched to subject 100 date 01/10/2021. I don't want this to
happen. Once subject 100 is used to match subject 1, I don't want any other
person from file 1 to be matched with subject 100.
I hope you can help me with this, I suspect, not uncommon matching problem!
Thank you,
John
John David Sorkin M.D., Ph.D.
Professor of Medicine, University of Maryland School of Medicine;
Associate Director for Biostatistics and Informatics, Baltimore VA Medical
Center Geriatrics Research, Education, and Clinical Center;
Former PI Biostatistics and Informatics Core, University of Maryland School of
Medicine Claude D. Pepper Older Americans Independence Center;
Senior Statistician University of Maryland Center for Vascular Research;
Division of Gerontology, Geriatrics and Palliative Medicine,
10 North Greene Street
GRECC (BT/18/GR)
Baltimore, MD 21201-1524
Cell phone 443-418-5382
______________________________________________
[email protected] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide https://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
______________________________________________
[email protected] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide https://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
