Extracting months from dates with atypical format (%tcdd-Mon-CCYY_HH:MM)

[Sergey Alexeev]

Dear Stata users,

I have an elementary problem that I somehow fail to solve. It's already been a few days of struggle.

I have patient-level data with hospital admission dates. For example,
12-Nov-2019 17:34

Stata acknowledges the date as shown by this in "Format" of the variable
%tcdd-Mon-CCYY_HH:MM

I want to include months fixed effects into my model to see what will happen. Thus I need to extra month from these dates. When I use the usual command, such as
gen monthly_date = mofd(hospAdmTime)
I get dots.


Here is 0.1 of my date variable.

* Example generated by -dataex-. To install: ssc install dataex
clear
input double hospAdmTime
    1898421420000
    1896442080000
    1.8888276e+12
1891769819999.998
1890307319999.998
1891146300000.002
    1901499480000
1903475999999.998
1888123560000.002
1890399300000.002
    1908565440000
1906909800000.002
    1890471660000
    1.8888516e+12
    1896013860000
1899059699999.998
1892028899999.998
    1905972780000
    1902095280000
1889199299999.998
1883128560000.002
    1893848940000
1898796300000.002
    1901622540000
1904807099999.998
    1.8903219e+12
1889264099999.998
1900870499999.998
    1.8857934e+12
    1887447660000
    1884433440000
    1900547760000
1903526459999.998
    1.8893646e+12
    1.8953304e+12
    1891432440000
end
format %tcdd-Mon-CCYY_HH:MM hospAdmTime

[Hemanshu Kumar]

At least two ways of doing it, pick the one you find more useful:

gen mm = clockpart(hospAdmTime,"month")
gen yy = clockpart(hospAdmTime,"year")

gen mmyy = mofd(dofc(hospAdmTime))
format %tm mmyy

[Sergey Alexeev]

Thank you, Hemanshu Kumar,

I can confirm that your code performs as expected.