Calculate date of birth from current date and age (in years, months and days)

[MH Sandoval]

Dear

I need help to calculate the date of birth from the age of people. I know the exact year, months and days that a person has until today (date_today). But, I need the dates of birth. I searched and there are several examples of how to calculate the age from the date of birth and a date of interest, but I did not find a reverse example (exact date of birth (year, day and month) from the exact age (age, months and days to current date).

Please, I appreciate any guidance.
the data variables are:

id: individual identifier
age_years: age in years to the present day
age_months: age in months to the present day
age_days: until today
date_today: reference date

A person who is 10 years, 0 months and 7 days old, means that he/she was born on February 12, 2015.
I sincerely hope you can help me

Best regards

Moses


copy starting from the next line -------------- --------

* Example generated by -dataex-. For more info, type help dataex
clear
input float id int age_years byte(age_months age_days) float date_today
1   9  3 27 23791
2  56 11 25 23791
3  55  3 20 23791
4  60  4 26 23791
5  59  8  0 23791
6  49  7  2 23791
7  50  0  3 23791
8  60  3 27 23791
9  63  8 11 23791
10  62  5 27 23791
11  49  4 12 23791
12  50  8  4 23791
13  65  9 28 23791
14  62  4 14 23791
15  61  1 15 23791
16  53  2 28 23791
17  56  3 24 23791
18  64  2  4 23791
19  60  4 21 23791
20  62  5 16 23791
21  43  5  5 23791
22  56 11 10 23791
23  61  4 10 23791
24  60  5 14 23791
25  44 11  7 23791
26  52  0 13 23791
27  60  5 26 23791
28  61  2 13 23791
29  60  7  5 23791
30  59 11 27 23791
31  48 11 11 23791
32  45  8 12 23791
33  60  2 15 23791
34  58  1 18 23791
35  61  4 14 23791
36  61  8  9 23791
37  48 11 22 23791
38  57  6  6 23791
39  58  9 19 23791
40  52  7 20 23791
41  54  7 28 23791
42  60  4 19 23791
43  61  3  3 23791
44  60  0 30 23791
45  60  0 23 23791
46  62  7 20 23791
47  58  7 13 23791
48  61  1 16 23791
49  64  1 23 23791
50  57  9  3 23791
51  59  0 29 23791
52  76 11  7 23791
53  62  0  9 23791
54  59  7 17 23791
55  50  3 29 23791
56  49  3 29 23791
57  57  4 10 23791
58  55  8  8 23791
59  60 10  2 23791
60 103  8 27 23791
61  61 11 28 23791
62  61  3 29 23791
63  47  1  0 23791
64  57  3 15 23791
65  54  3 11 23791
66  61 10 26 23791
67  56  6  6 23791
68  50  5  4 23791
69  53  4 13 23791
70  62  8 27 23791
71  60  9 20 23791
72  45  1 17 23791
73  48  4 28 23791
74  60  9 24 23791
75  60  3 28 23791
76  57  6 17 23791
77  61 10 17 23791
78  56  7 13 23791
79  58 11  7 23791
80  17  0 18 23791
81  58 10 14 23791
82  61  1  8 23791
83  44  9 10 23791
84  51  1  8 23791
85  51  9 28 23791
86  70  4 26 23791
87  58  9 30 23791
88  53  2 15 23791
89  43  5 13 23791
90  60  7 14 23791
91  60  0 14 23791
92  52  0  6 23791
93  61  5 15 23791
94  59 10  1 23791
95  57  5  3 23791
96  48  3  2 23791
97  59  4  1 23791
98  59 11 15 23791
99  45 10 18 23791
100  57 11 14 23791
end
format %td date_today

copy up to and including the previous line --------- --------

[Nick Cox]

There is a bundle of relevant functions in Stata 18, which is presumed to be what you're using unless you state otherwise.

If you are using some other version, please tell us what it is, so that people can try to gauge what is accessible to you.

[MH Sandoval]

Thank you Nick
I am currently using stata 16 and have tried to turn it around by generating variables but got lost in the attempt.
I appreciate any guidance

[Hemanshu Kumar]

This might be one way to do it:

clear
input float id int age_years byte(age_months age_days) float date_today
1 0 0 1 23791
2 2 1 0 23791
3 10 0 7 23791
end
format %td date_today

gen int age_in_months = age_years * 12 + age_months
gen double mm = dofm(mofd(date_today) - age_in_months)
gen double this_day_that_month = mdy(month(mm), day(date_today), year(mm))
gen double dob = this_day_that_month - age_days
drop age_in_months mm this_day_that_month
format %td dob

which produces:

. list, noobs abbrev(20)

  +-----------------------------------------------------------------+
  | id   age_years   age_months   age_days   date_today         dob |
  |-----------------------------------------------------------------|
  |  1           0            0          1    19feb2025   18feb2025 |
  |  2           2            1          0    19feb2025   19jan2023 |
  |  3          10            0          7    19feb2025   12feb2015 |
  +-----------------------------------------------------------------+

[MH Sandoval]

Dear Hemanshu
Thank you very much for the suggestion, I'll see how it works!!!

[MH Sandoval]

Dear Hemanshu,

To test the suggestion I used it in a database in which I know the date of birth.
The results obtained vary by a few days (1, 2, 3 or more) in some cases.
The older the person is the more difference of days there is.
I think it must be associated with leap years (February with 29 days).
Even in some cases, when the date is close to the change of month it also generates some differences.
However, since there is no other way, working with a difference of a few days seems not so problematic.
Thank you very much for your help

[Hemanshu Kumar]

That's strange, I thought the calculation would be exact. Could you show me a data extract that includes some of the problem cases? (please include the known date of birth)

[MH Sandoval]

I recalculated the difference and it is 2 to 3 days. As I mentioned it is not so problematic for those who stay in the same month but in some cases it changes month.
birth_original is the variable with the actual date (new database).




copy starting from the next line --------------------- -

* Example generated by -dataex-. For more info, type help dataex
clear
input int age_years byte(age_months age_days) float(date_today birth_original)    int    dob
56 11 25 23791  2977  2975
55  3 20 23791  3592  3590
59  8  0 23791  1998  1996
60  3 27 23791  1759  1757
63  8 11 23791   526   524
49  4 12 23791  5761  5758
50  8  4 23791  5281  5279
62  4 14 23791  1011  1008
53  2 28 23791  4345  4342
56  3 24 23791  3223  3221
64  2  4 23791   352   349
60  4 21 23791  1735  1732
62  5 16 23791   978   976
56 11 10 23791  2993  2990
61  4 10 23791  1380  1377
60  5 14 23791  1711  1709
44 11  7 23791  7379  7376
61  2 13 23791  1438  1435
60  7  5 23791  1659  1656
59 11 27 23791  1880  1877
61  4 14 23791  1376  1373
61  8  9 23791  1258  1256
57  6  6 23791  2784  2781
58  9 19 23791  2314  2311
61  3  3 23791  1417  1415
62  7 20 23791   912   910
76 11  7 23791 -4309 -4312
59  7 17 23791  2012  2009
50  3 29 23791  5409  5407
49  3 29 23791  5774  5772
60 10  2 23791  1570  1568
61 11 28 23791  1148  1145
61  3 29 23791  1391  1389
57  3 15 23791  2866  2864
54  3 11 23791  3966  3964
61 10 26 23791  1180  1178
56  6  6 23791  3150  3147
50  5  4 23791  5373  5371
53  4 13 23791  4299  4296
60  9 20 23791  1583  1580
60  9 24 23791  1579  1576
57  6 17 23791  2773  2770
56  7 13 23791  3112  3109
58 11  7 23791  2265  2262
51  9 28 23791  4862  4859
70  4 26 23791 -1923 -1926
58  9 30 23791  2303  2300
53  2 15 23791  4358  4355
43  5 13 23791  7921  7919
61  5 15 23791  1344  1342
59 10  1 23791  1936  1934
59  4  1 23791  2120  2117
59 11 15 23791  1892  1889
57 11 14 23791  2623  2620
48  7 19 23791  6028  6025
53  6  5 23791  4246  4243
58  6  3 23791  2422  2419
58  8  2 23791  2361  2359
52  2 25 23791  4714  4711
53  9 29 23791  4130  4127
58  8 15 23791  2348  2346
50  7 18 23791  5298  5295
60  2  6 23791  1811  1808
58 10 22 23791  2280  2278
62 10 29 23791   812   810
60 11 14 23791  1528  1525
60  8  5 23791  1628  1626
58  7 23 23791  2371  2368
58  7 18 23791  2376  2373
60  4 24 23791  1732  1729
58  9 27 23791  2306  2303
53  7 20 23791  4200  4197
60  9 24 23791  1579  1576
57  9 12 23791  2686  2683
60 10  8 23791  1564  1562
53 10 15 23791  4113  4111
60  6 27 23791  1668  1665
51  5 22 23791  4990  4988
59  6 12 23791  2048  2045
55  8  3 23791  3455  3454
57  6 16 23791  2774  2771
60  4 11 23791  1745  1742
51  7 12 23791  4939  4936
59  3 27 23791  2124  2122
60  3 23 23791  1763  1761
59  6  9 23791  2051  2048
58  4 29 23791  2457  2454
60  9 24 23791  1579  1576
57  4 25 23791  2826  2823
55  4  0 23791  3582  3579
58  2  1 23791  2546  2543
59  6 21 23791  2039  2036
59  9 20 23791  1948  1945
61  5  9 23791  1350  1348
62  2 21 23791  1065  1062
59  9  2 23791  1966  1963
59  8  0 23791  1998  1996
61  9  1 23791  1236  1233
58  2 29 23791  2518  2515
60  3  9 23791  1777  1775
end
format %td date_today
format %td birth_original
format %td dob

copy up to and including the previous line ---------------- -

[Hemanshu Kumar]

I don't think this is an error with the way I am constructing the date of birth. There seems to be some issue with your original data. Take the third observation, for example, who is 59 years, 8 months, and 0 days old. If today's date is 19 February 2025, clearly they must have been born on the 19th (since they are exactly 59 years and 8 months old). But as per your birth_original, they were born on 21 June 1965.

[MH Sandoval]

You are right. I will check the data.
Thank you very much.