Hello,
I have a panel data-set that I am in the process of cleaning. This data-set records violent incidents of patients in a mental health clinic. A patient can have different admissions and admissions can have multiple violent incidents. Admission ID also have its corresponding admission date.
I need to fix missing values for the admission Id and the Admission date. I can deduce the admission ID and Date by looking at previous and post incidents and the admission date variable. For example patient 2, have a missing admission ID and date for incident 9, but incident 9 happened between between 11/13/13 and 9/21/14, and these two date have the same admission date. Therefore incident 9 should be assigned admission id 51 and its corresponding admission date 9/1/2010.
For patient ID 1 there is no way to tell what is the correct value for missing value so it should remain as missing.
Below you can find a dataex sample.
Thank you,
Marvin
I have a panel data-set that I am in the process of cleaning. This data-set records violent incidents of patients in a mental health clinic. A patient can have different admissions and admissions can have multiple violent incidents. Admission ID also have its corresponding admission date.
I need to fix missing values for the admission Id and the Admission date. I can deduce the admission ID and Date by looking at previous and post incidents and the admission date variable. For example patient 2, have a missing admission ID and date for incident 9, but incident 9 happened between between 11/13/13 and 9/21/14, and these two date have the same admission date. Therefore incident 9 should be assigned admission id 51 and its corresponding admission date 9/1/2010.
For patient ID 1 there is no way to tell what is the correct value for missing value so it should remain as missing.
| Patient_ID | Admission_ID | Incident_ID | Incident_Date | Admission_Date |
| 1 | 1 | 10/4/2013 | ||
| 1 | 10 | 2 | 3/22/2014 | 11/7/2010 17:16 |
| 2 | 51 | 3 | 1/3/2013 | 9/1/2010 17:32 |
| 2 | 51 | 4 | 3/13/2013 | 9/1/2010 17:32 |
| 2 | 51 | 5 | 7/19/2013 | 9/1/2010 17:32 |
| 2 | 51 | 6 | 9/6/2013 | 9/1/2010 17:32 |
| 2 | 51 | 7 | 10/20/2013 | 9/1/2010 17:32 |
| 2 | 51 | 8 | 11/13/2013 | 9/1/2010 17:32 |
| 2 | 9 | 2/7/2014 | ||
| 2 | 51 | 10 | 9/21/2014 | 9/1/2010 17:32 |
| 2 | 5 | 11 | 9/27/2014 | 9/1/2010 17:32 |
| 2 | 51 | 12 | 4/29/2015 | 9/1/2010 17:32 |
| 2 | 51 | 13 | 9/26/2015 | 9/1/2010 17:32 |
Below you can find a dataex sample.
Code:
* Example generated by -dataex-. To install: ssc install dataex clear input byte(patient_id admission_id incident_id) double(incident_date_t admission_date_t) 1 . 1 19635 . 1 10 2 19804 1604769360000 2 51 3 19361 1598981520000 2 51 4 19430 1598981520000 2 51 5 19558 1598981520000 2 51 6 19607 1598981520000 2 51 7 19651 1598981520000 2 51 8 19675 1598981520000 2 . 9 19761 . 2 51 10 19987 1598981520000 2 5 11 19993 1598981520000 2 51 12 20207 1598981520000 2 51 13 20357 1598981520000 4 55 17 19283 1.6640664e+12 4 55 18 19289 1.6640664e+12 4 55 19 19289 1.6640664e+12 4 60 20 19563 1673448720000 4 61 21 19738 1698539280000 4 . 22 19743 . 4 61 23 19749 1698539280000 end format %td incident_date_t format %tc admission_date_t
Thank you,
Marvin