Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down on April 23, and its replacement, **statalist.org** is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Richard Williams <richardwilliams.ndu@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: Is there an -mi drop- command? |

Date |
Thu, 11 Apr 2013 21:29:17 -0500 |

At 11:38 AM 4/9/2013, you wrote:

I'm pretty sure the answer to this question is no: I've scoured theMI manual and can't find it.Let me explain what I'm looking for and why I think such a commandwould be useful.Working with a data set that has a great deal of missinginformation, I sometimes end up over-generating imputations. Thatis, my initial estimate of the number of imputations I will needends up too high. It would be nice to be able to tell Stata tosimply drop some of them, rather than having to start over andgenerate a smaller number. You might ask what the harm is inkeeping them all. Well, the more imputations you have, the longereach MI analysis takes, and if there are many analyses, the timestarts to add up.There is another reason I would like to see a command likethis. When the data include difficult patterns of missingness, suchas pairs of variables V1 and V2 that are never both non-missing inthe same observation (so their imputations are entirely dependent onyet other variables and there is no direct information about theirassociations), my experience has been that the imputed data setssometimes contain imputed values that are widely outside the rangeof the observed data, even by a few orders of magnitude. MIanalyses that include these data sets sometimes produce results thatare clearly absurd.And there is a third situation. When doing chained imputations,there is the issue of the burn-in. Again, we have to make aninitial guess and see what happens. Sometimes, I will generate alarge number of imputations using a long burn-in because earlierattempts with a shorter burn-in don't converge to stableestimates. Maybe I have 75 imputations after a burn-in of 30. If Istudy the convergence carefully, I may find that there are 3 or 4imputations where convergence has not been achieved, but the restlook OK. It would be nice to be able to tell Stata to drop those 3or 4--after all, 75 is just a round number and in most circumstances71 or 72 imputations will serve nearly as well.While the -mi estimate- command itself permits specification of anumlist with the numbers of the imputations I want to use, having torepeatedly specify this list is a nuisance. It would be simpler tobe able to drop those imputations, re-save the data set, and thenmove on to analysis.The most cogent objection I can think of to this proposal is thatperhaps I shouldn't be using MI at all in these situations, or,perhaps, that my imputation models need some serious re-thinkingwhen the imputations show this kind of irregularity.Any thoughts? Clyde Schechter Dept. of Family & Social Medicine Albert Einstein College of Medicine Bronx, NY, USA * * 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/

------------------------------------------- Richard Williams, Notre Dame Dept of Sociology OFFICE: (574)631-6668, (574)631-6463 HOME: (574)289-5227 EMAIL: Richard.A.Williams.5@ND.Edu WWW: http://www.nd.edu/~rwilliam * * 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/

**References**:**st: Is there an -mi drop- command?***From:*Clyde B Schechter <clyde.schechter@einstein.yu.edu>

- Prev by Date:
**st: comparing biprobit marginal effects CIs** - Next by Date:
**st: Hausman test, rank of the differenced variance matrix, (V_b-V_B is not positive definite)** - Previous by thread:
**st: Is there an -mi drop- command?** - Next by thread:
**Re: st: Is there an -mi drop- command?** - Index(es):