Hi,
I have two really large datasets (5,00,000 observations each) and I have been using the command "matchit id allNames using "xyz.dta", idusing(familyid) txtusing(allNamesFamily)" to match names between these two datasets.
However, matchit is taking a really really long time to carry out the fuzzy match (almost 24 hours). I have decided to run the same command but on smaller groups now however I am not sure how to create a loop function for it.
Essentially,
I want STATA to create district-level groups, carry out the above matchit command for each group, and save all the results together. The idea is that matchit only has to look for searches within each of these district-level sub-groups and not the whole dataset.
Can anyone help me with writing such a loop?
Related Posts with How to create a loop function to implement fuzzy matching on a large dataset?
panel data regression Hey, I'm running panel data regressions, but I want to find out the coefficients for each country …
Descriptive Data Table (Socio-demographic data)Hi there, I was wondering whether I could get a table like this on STATA? Array Thanks, Aslan …
Problem (omitted because of collinearity Code: probit checking_account lending_interest i.size c.firm_age i.Legal_Status i.sector i.gender …
Generating a Count Variable using the 90 percentile centered on a five day window Hi I am working with a panel data set for 300 districts from 1959 to 1999 using Stata 13. I have go…
Cem (coarsened exact matching) create identifier for matched pairsDear Statalist Users, My data is on the district-year level and I am looking at the effect of mine …
Subscribe to:
Post Comments (Atom)
0 Response to How to create a loop function to implement fuzzy matching on a large dataset?
Post a Comment