The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Kenya » Under-5, Infant and Neonatal Mortality Rates
Under-5, Infant and Neonatal Mortality Rates [message #27860] Fri, 13 October 2023 12:02 Go to next message
sokiya is currently offline  sokiya
Messages: 80
Registered: May 2017
Location: Nairobi
Senior Member
Hi,

I am using the attached do file from DHS GitHub repository to reproduce the results in the report. The do file produces the attached dta file. I am wondering what's the rule of thumb in selecting the correct estimates where covariate is "All" since there are 4 different results. is it the run number? Is it advisable to comment out sections that estimate the rates that am not interested in since the do files takes about 15 minutes to run?

Thank you!
Re: Under-5, Infant and Neonatal Mortality Rates [message #27864 is a reply to message #27860] Sat, 14 October 2023 04:07 Go to previous messageGo to next message
sokiya is currently offline  sokiya
Messages: 80
Registered: May 2017
Location: Nairobi
Senior Member
Please disregard my question. I realize the windows (lw and uw) are different so the correct one should be 5 years preceding the survey.
Re: Under-5, Infant and Neonatal Mortality Rates [message #27876 is a reply to message #27864] Mon, 16 October 2023 08:46 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3208
Registered: February 2013
Senior Member
Following is a response from Senior DHS staff member, Tom Pullum:

You are running the program with the maximum number of options--that's why it is taking so long. The difference between the different runs with "All" is in the reference period. Here is a listing that includes lw and uw:


. list lw uw *MR if label=="All", table clean

lw uw NNMR PNNMR IMR CMR U5MR
1. -14 -10 20.59 13.62 34.20 12.95 46.71
2. -9 -5 24.84 13.03 37.87 8.26 45.82
3. -9 0 22.96 12.09 35.04 8.65 43.40
74. -4 0 21.23 11.23 32.46 9.02 41.18


Here lw and uw identify the lower and upper ends of the "window" of deaths and exposure. Thus, (-14,-10) means 10-14 years ago. (-9,-5) is 5-9 years ago. (-9,0) is 0-9 years ago, that is, the past 10 years. (-4,0) is the past 5 years. The program (which I wrote) give you a lot of flexibility in the specification of time intervals and covariates, but you may not need that much. Let us know if you have other questions.

[Updated on: Mon, 16 October 2023 08:47]

Report message to a moderator

Previous Topic: Missing Data on Health Insurance
Next Topic: Construction of covariates
Goto Forum:
  


Current Time: Wed Dec 11 18:34:46 Coordinated Universal Time 2024