Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: st: Reconciling Quantile Regression Results
From
"Santos Silva, J.M.C." <[email protected]>
To
<[email protected]>
Subject
RE: st: Reconciling Quantile Regression Results
Date
Thu, 6 Feb 2014 10:34:46 GMT
Dear Bob,
My guess is that you are using Stata 13 at home and an earlier version
in the office. (Stata 13 has a different way of computing standard errors
for qreg.) You should be able to replicate the office results at home using
the -version- command; the other way around is more difficult, but if you
want to compute robust standard errors you can try qreg2 instead of
qreg ;-)
All the best,
Joao
> Dear Statalist
>
>
> I have a curious problem. I am using quantile regression to estimate
> the betas and standard errors of a data set with five implicates for
> each household. Using the same do file and data file ( both on a thumb
> drive) I get different results on two different computers, home and
> office. The estimates for the beta cofficients for both are
> essentially the same while the standard errors are wildly different.
> (I have attached the results from each computer below using the same
> do file.) Obviously, I have tried to control for everything, but have
> found nothing to answer the discrepancy. I have located the
> discrepancy to the following step within the longer do file.
>
>
> statsby _se if (age<66), by (imp) : qreg networthK ZAgeCl1 ZAgeCl2
> ZAgeCl3 ZAgeCl4 ZWhite ZKids ZNormalYK ZDidSave ZCreditGood
> ZLegacyImp ZLRPlanHor ZLifeRemain ZLatePay ZUnemployHouse ZBankruptcy
> ZHomeOwner hprim_mort ZHasOthReal ZHasStocks ZHasBus ZRiskTaker
> ZHasEducDebt ZRecInherit ZExpInheritValK ZFinLucky, quantile (.90)
>
>
> However, the same command generates similar (virtually identical)
> betas and different standard errors.
>
> Below are the results from one computer and then another.
>
> Office Results:
>
> varname beta sigma tstat signf
> ------------------------------
> -------------------------------
> 1. _ZAgeCl1 -49.26508 13.56417 -3.632001 ***
> 2. _ZAgeCl2 -59.56253 11.21504 -5.310952 ***
> 3. _ZAgeCl3 -66.35434 10.08387 -6.580249 ***
> 4. _ZAgeCl4 -53.12152 10.45471 -5.081109 ***
> 5. _ZWhite -10.01998 5.160771 -1.941567 *
> -------------------------------------------------------------
> 6. _ZKids -2.201806 2.071385 -1.062963
> 7. _ZNormalYK 23.00146 .7403392 31.06881 ***
> 8. _ZDidSave -13.60232 5.286685 -2.57294 **
> 9. _ZCreditGood 4.081276 5.569656 .7327698
> 10. _ZLegacyImp 5.381646 4.49124 1.198254
> -------------------------------------------------------------
> 11. _ZLRPlanHor -9.36632 5.308161 -1.764513 *
> 12. _ZLifeRemain .1218808 .1869905 .6518021
> 13. _ZLatePay 4.8959 10.39511 .4709811
> 14. _ZUnemployHouse 7.037433 6.363559 1.105896
> 15. _ZBankruptcy 6.262784 6.837511 .9159451
> -------------------------------------------------------------
> 16. _ZHomeOwner 23.97475 14.21518 1.686559 *
> 17. _hprim_mort -69.95344 14.28528 -4.896891 ***
> 18. _ZHasOthReal 81.05359 38.13787 2.125278 **
> 19. _ZHasStocks 4.804673 7.735104 .6211517
> 20. _ZHasBus 396.266 21.40136 18.51593 ***
> -------------------------------------------------------------
> 21. _ZRiskTaker 9.442237 6.313608 1.495537
> 22. _ZHasEducDebt -16.83249 6.137966 -2.742357 ***
> 23. _ZRecInherit .5936526 6.310728 .0940704
> 24. _ZExpInheritValK 1.586981 .0537434 29.52886 ***
> 25. _ZFinLucky .2072812 5.082915 .04078
> -------------------------------------------------------------
> 26. _cons 31.75737 12.20822 2.60131 ***
>
>
> Home Results
>
> varname beta sigma tstat signf
> ------------------------------
> -------------------------------
> 1. _ZAgeCl1 -49.26507 286.9954 -.171658
> 2. _ZAgeCl2 -59.56252 206.6773 -.2881909
> 3. _ZAgeCl3 -66.35435 169.4279 -.3916376
> 4. _ZAgeCl4 -53.12151 140.72111 -.377495
> 5. _ZWhite -10.01998 119.2511 -.040242
> -------------------------------------------------------------
> 6. _ZKids -2.201802 41.99823 -.0524261
> 7. _ZNormalYK 23.00146 .7515529 30.06524 ***
> 8. _ZDidSave -13.60232 105.5748 -.1288406
> 9. _ZCreditGood 4.081271 120.856 .0337697
> 10. _ZLegacyImp 5.381646 98.09854 .0548596
> -------------------------------------------------------------
> 11. _ZLRPlanHor -9.36616 104.1479 -.0899328
> 12. _ZLifeRemain .1218808 4.063498 .029994
> 13. _ZLatePay 4.895901 228.7711 .0214008
> 14. _ZUnemployHouse 7.037432 143.9856 .048876
> 15. _ZBankruptcy 6.262784 157.4319 .03978091
> -------------------------------------------------------------
> 16. _ZHomeOwner 23.97475 172.3645 .1390934
> 17. _hprim_mort -69.95345 136.6182 -.512036
> 18. _ZHasOthReal 81.05359 127.5272 -.6355788
> 19. _ZHasStocks 4.804671 120.1264 .0399968
> 20. _ZHasBus 396.266 122.9045 3.224178 ***
> -------------------------------------------------------------
> 21. _ZRiskTaker 9.442244 112.3682 .0840295
> 22. _ZHasEducDebt -16.83249 140.7864 -.1195605
> 23. _ZRecInherit .5936565 119.3811 .0049728
> 24. _ZExpInheritValK 1.586981 .1220853 12.92275 ***
> 25. _ZFinLucky .2072808 115.2274 .0017989
> -------------------------------------------------------------
> 26. _cons 31.75737 223.852
>
>
> Any ideas? I am perplexed.
>
> Bob
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/