Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: gnbreg problem with replicate weights

From   "Tim Brown" <>
To   <>
Subject   st: gnbreg problem with replicate weights
Date   Mon, 12 Feb 2007 15:00:46 -0800


I am running gnbreg using the svy command and using replicate weights.  I
find the I get results when I do not use the lnalpha option, but when I
attempt to use this option (the whole reason for using gnbreg), Stata will
hang up partially way through the replicates weights and never produce
estimates (we let it run over night to be sure we weren't giving up too
quickly).  Here is my code:

svyset [pweight=fwgt], jkrweight(fwgt1-fwgt51, multiplier(.98))
vce(jackknife) mse

global xiscounty1 "female nonwhite age2534 age3544 age4554 age5564 age6574
age75plus married highschoolgrad somecollege collegegrad postgrad year02
lnpscicounty lnpercentrelcounty medianti01 medianti_county
lnpricepack_county county_dums2-county_dums58 income_dum1-income_dum5
quarter1 nonwh_lnrelsc nonwh_lnsc cntymed_lnsc cntymed_lnrelsc nonwh_cntymed
nonwh_lnsc_cntymed nonwh_lnrelsc_cntymed"

global county_dums2 "county_dums3-county_dums24 county_dums26-county_dums45

svy: gnbreg numcigs30d $xiscounty1, lnalpha($county_dums2)

(running gnbreg on estimation sample)

Jackknife replications (51)
----+--- 1 ---+--- 2 ---+--- 3 ---+--- 4 ---+--- 5 

This is as far as it gets.

Am I doing something wrong?



Timothy T. Brown, PhD
Associate Director of Research
Petris Center on Health Care Markets and Consumer Welfare 
University of California at Berkeley 

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index