Hello,
I am attempting to recreate Medical Expenditure Panel Survey expenditure data for Children with Special Health Care Needs "Chevarley FM. Utilization and Expenditures for Children with Special Health Care Needs. Research Findings No. 24. January 2006. Agency for Healthcare Research and Quality, Rockville, Md. http:www.meps.ahrq.govrf24rf24.pdf”
The median values in the report are reported to 2 decimal places, however when I run my code
using the svy statement the output produces whole numbers. For example, Table 4 of the Chevarley paper reports a median per child with expense of $316.29 for all children, but the Stata output is 317. 'ANYEXP' is a variable for if the child has any expenditure. I used the epctile command after reviewing previous StataList posts.
svyset [pweight=SQPQW00F], strata(VARSTR) psu(VARPSU) singleunit(centered)
epctile TOTEXP if ANYEXP==1, svy percentiles(50) subpop(CHILD00)
Does anyone know how to get a more precise estimate or know of a better method to obtain median values?
The MEPS data files I am using are available free here https://meps.ahrq.gov/mepsweb/data_s...fNumber=HC-050
thanks!
I am attempting to recreate Medical Expenditure Panel Survey expenditure data for Children with Special Health Care Needs "Chevarley FM. Utilization and Expenditures for Children with Special Health Care Needs. Research Findings No. 24. January 2006. Agency for Healthcare Research and Quality, Rockville, Md. http:www.meps.ahrq.govrf24rf24.pdf”
The median values in the report are reported to 2 decimal places, however when I run my code
using the svy statement the output produces whole numbers. For example, Table 4 of the Chevarley paper reports a median per child with expense of $316.29 for all children, but the Stata output is 317. 'ANYEXP' is a variable for if the child has any expenditure. I used the epctile command after reviewing previous StataList posts.
svyset [pweight=SQPQW00F], strata(VARSTR) psu(VARPSU) singleunit(centered)
epctile TOTEXP if ANYEXP==1, svy percentiles(50) subpop(CHILD00)
Does anyone know how to get a more precise estimate or know of a better method to obtain median values?
The MEPS data files I am using are available free here https://meps.ahrq.gov/mepsweb/data_s...fNumber=HC-050
thanks!
Comment