On Fri, 20 Dec 2002, Kevin Timothy Arceneaux wrote:
> Prof. King,
>
> My understanding is that interactions, which will be used in the analysis,
> should be included in the imputation data set. Say I have missing values
> in X1 and X2, which produce missing values in X1*X2. Amelia will impute
> values for X1, X2, and X1*X2, but it will not necessarily be the case that
> the imputed values for X1*X2 will equal the multiplication of X1 and X2.
> Is there a way to set this sort of constraint in Amelia? (I am using the
> Windows version).
>
> If not, would it be appropriate to set the missing values in X1*X2 to some
> arbitrary number (e.g., -9) and then recomputed the interaction with the
> imputed values for each of the resulting imputed data sets? Or should I
> allow Amelia to impute missing values in X1*X2, and then correct these
> values in the resulting imputed data sets?
Good question. I would alow Amelia to impute x1*x2 and then recompute
them from the contituent (imputed) variables. that allows the interaction
to have an effect and it produces consistent estimates.
One issue in Amelia (and all methods of imputation) is that if x1 is
missing for a variable but x2 is observed, x1*x2 will be coded as missing
even though it is partially observed. nothing you can do about this tho.
Gary
: Gary King, King(a)Harvard.Edu http://GKing.Harvard.Edu :
: Center for Basic Research Direct (617) 495-2027 :
: in the Social Sciences Assistant (617) 495-9271 :
: 34 Kirkland Street, Rm. 2 HU-MIT DC (617) 495-4734 :
: Harvard U, Cambridge, MA 02138 eFax (928) 832-7022 :
>
> Thanks for your help,
> Kevin.
>
>
> Kevin Arceneaux
> Ph.D. Candidate
> Department of Political Science
> Rice University
> 6100 Main St., MS 24
> Houston, TX 77005
> (713) 348-2107
>
Dear Sir/Madam,
My name is Peter Severa, I am Business Development Director
at E-mailPromo Company. Your name was at a business-related
website. Do I understand correctly that your activity is
related to business? May I know, what industry are you in,
specifically?
If you may be interested in new customers, our company will
be glad to assist you with promotion of your business. We offer
databases of email addresses for direct email advertising,
we deliver mailing campaigns, offer bulk-email web hosting
for advertising projects. Also, we can design an attractive
html message and a supporting website.
Today, our database includes over 50 million valid and original
verified e-mail addresses. It is constantly being updated with
addresses from all over the world. We can also create
customized databases targeted at your request.
By the way, if you would like us to remove you from our
database, please send a message to info(a)e-mailpromo.net with
"Remove" in the Subject line.
Usually, mailing of 1 million addresses allows to brings
1000-5000 new visitors interested in your offer to your site.
Your advertising campaign will be fully legal. It will
include a remove instruction, thus it will be in compliance
with the new e-mail bill section 301. Under Bill S. 1618
TITLE III passed by the 105th US Congress.
We use our own mailing software which can be ideally adjusted
for every customer. We have a high-bandwidth channel and a high
power server.
The constant demand for our service allows us to keep low
prices.
Please see all details, FAQ about our service, and references
of our customers at:
http://www.e-mailpromo.net
If you have any questions or a request, please do _not_ reply me
just with a Reply button.
I will be happy to answer every your question.
Thank you,
Peter Severa
Business Development Director
E-mailPromo Company
www.e-mailpromo.net
I am trying to use the time series option in the windows version of Amelia (I was able to read in the data successfully as text after the initial problems). How do I specify the interperiod length (in the manual, it is stated that you type 1 if your data are yearly) when my data are monthly in defining the AMtstep?
Thanks for your help.
Eric Akunda.
I am trying to read a data set into Amelia (I am a first time user!) but
failing.
My computer runs Windows 2000 Professional. I have saved the data as a
Microsoft Excel 2.1 file named ssri.xls. The data set has 20 variables
with a total of 498 observations that are a pooled data set of five
different products.
When I try to input the file I type C:\ssri.xls which resides in the C:\
drive noting the initial problems I had experienced with even trying to
run Amelia itself from folders with longer names (at least I ran the
example with sample.asc and it worked!). When I type enter, I get a
message that reads as follows:
C:\DOCUM~1\Eric\LOCALS~1\Temp\dload$$$.fmt' exit status 3
Error: Program was unable to read file contents. Please check
instructions to make sure input is of correct type.
I know this may have happened to someone else but having gone through
the threads in the archives of the listserv, I was not able to see this
addressed (and did not find it in the FAQs). Is it a problem that's
unique to me?
Thanks for your time and answering the query.
Eric.
Eric A. Akunda
Doctoral Student (Marketing)
Kenan-Flagler Business School
University of North Carolina - Chapel Hill
CB # 3490 McColl
Chapel Hill, NC 27599-3490
Tel: 919-962-0783
Voice: 919-918-1222
Fax: 919-962-7186
Email: eric_akunda(a)unc.edu <mailto:eric_akunda@unc.edu>
On Sat, 7 Dec 2002, Diogo L Pinheiro wrote:
> Dear Dr. King, I have a 55x17 TSCS data set with about 40 variables. The
> model I intend to run afterwards involves using dummy variables for
> countries and time. When I use Amelia with all these dummies, the
> program will disappear a few seconds after I hit run (i get the screen
> Amelia Begins, and then it all disapears). I've tried including the
> dummies both manually and by setting amusecs=1 and amusets=1 and setting
> it as a nominal variable. I've got a AMD 1.2 Ghz computer with 256 mb of
> RAM. Is ther anything I can do? (I've set the max_workspace do different
> values, from 100 do 200, and I always get the same.)
if you're using the features for TSCS variables, then you may have an
enormous number of variables. the number of parmaters in Amelia goes up
fast in the number of variables, and so this may be the problem. I
suggest (in addition to counting exactly how many variables you are
implicitly using) you try it with a small number of variables, get that to
work, and then gradually increase.
> Also, another problem that I often run into is that sometimes I will set
> Amfully correctly, and when I start to run Amelia, I get a message
> saying that Amfully is not properly defined. After I get that message I
> go back to the AMfully screen and for some reason the values there are
> different from the ones I entered. I found that that happens to most
> fields after getting the error message, and so I have to reset all the
> values. Are there any solutions for that? By the way, I use the windows
> version. Thanks Diogo Pinheiro
>
Amelia for windows works by checking all the global values and resetting
any that are invalid. So that's a feature, or at least it was by design,
and not really a bug. Tho I see your point..
Gary
: Gary King, King(a)Harvard.Edu http://GKing.Harvard.Edu :
: Center for Basic Research Direct (617) 495-2027 :
: in the Social Sciences Assistant (617) 495-9271 :
: 34 Kirkland Street, Rm. 2 HU-MIT DC (617) 495-4734 :
: Harvard U, Cambridge, MA 02138 eFax (928) 832-7022 :
I'd take the square root or log-plus-one to transform your data to a more
normal scale before Amelia and then untransform after.
otherwise, the time-series-cross-sectional aspect will be the most
difficult. I would use Amelia's special features for these types of data,
but then you should carefully check the imputations, as it seems you have
been doing, to make sure they're reasonable. Amelia should work in these
situations a good deal better than SAS because of these features, but it
will not necessarily fix all the problems.
Best of luck with your research,
Gary King
: Gary King, King(a)Harvard.Edu http://GKing.Harvard.Edu :
: Center for Basic Research Direct (617) 495-2027 :
: in the Social Sciences Assistant (617) 495-9271 :
: 34 Kirkland Street, Rm. 2 HU-MIT DC (617) 495-4734 :
: Harvard U, Cambridge, MA 02138 eFax (928) 832-7022 :
On Mon, 2 Dec 2002, Akunda, Eric wrote:
> Dear Prof King,
>
> I have a data set that has missing values for some variables in the
> following way. I would like to describe the missing data to determine
> whether using Amelia will be appropriate (I have read the article that
> accompanies Amelia).
>
> My data are for the pharmaceutical industry. I have # of prescriptions
> sold as my response (which I convert to market shares). I then have as
> covariates the # of minutes the salespeople detail to physicians, the
> counts of these details, the detailing expenditures (in dollars), the
> price per prescription in dollars, the expenditure on direct-to-consumer
> advertising, the events and meetings expenditures (in dollars), and
> journal advertising. The latter three variables are incomplete because
> the firm collecting the data (from whom we purchased the same) did not
> collect the data at the time, and seems to have began to collect these
> variables on enquiries from its clients. The data are time series
> (monthly) and cross-sectional (by brand), and based on the fact that the
> brands enter the market sequentially create and unbalanced panel. I
> intend to use Amelia (the time-series cross-sectional function) to
> impute the missing values. However, my question is, would this be
> appropriate, based on the fact that the data are not really missing at
> random? Would this necessitate coming up with a specification unique to
> the data as you have advised in the paper? Eliminating observations in
> this dataset is unthinkable because the observations are very few (about
> 500 for the pooled brands including those with missing values on the
> three covariates). I have tried to impute the data using proc MI in sas
> (based on the book by Paul Allison), but some of the imputed values were
> poor. I have the feeling that the importance resampling coupled with
> specifying a range for the missing values would improve the imputations
> (especially so because the specification for the panel data is clearer
> compared to the proc mi specification based on "my" understanding of
> both). What would your advice be? Thanks very much in advance for taking
> your precious time to provide an answer. I will update you on my
> experiences with using Amelia for this dataset.
>
> Kindest regards,
> Sincerely,
> Eric Akunda.
> PhD student, Marketing.
> University of North Carolina at Chapel Hill.
>
Clarify is now Amelia-ready in the sense that there are facilities within
it to use multiply imputed data sets. So you don't need other software
any longer. For details of how to use this, see
http://gking.harvard.edu/clarify/docs/node26.html
Thanks for your interest.
Gary King
: Gary King, King(a)Harvard.Edu http://GKing.Harvard.Edu :
: Center for Basic Research Direct (617) 495-2027 :
: in the Social Sciences Assistant (617) 495-9271 :
: 34 Kirkland Street, Rm. 2 HU-MIT DC (617) 495-4734 :
: Harvard U, Cambridge, MA 02138 eFax (928) 832-7022 :
On Mon, 2 Dec 2002 TxLiberal(a)aol.com wrote:
> I am in a Mulitvariate class at the University of Houston. We are using
> Stata, Clarigy and Amelia and were told to use the MIEST command in
> Stata. However, we cannot find where to download the MIEST software.
> Any assistance you could offer would be greatly appreciated.
>
> Thanks,
>
> Tom
>
We offer you e-mail addresses databases for advertisement
mailing; we sell databases also carry out mailing and hosting
for the advertising projects.
We can work on a turnkey project create a site with original design,
program and subject contents. Our databases updates constantly
with the e-mail addresses from all over the world.
Their validity and originality are verified. Today they contains over
than 50 million addresses.
We use our own mailing soft which can be ideally adjusted for every
customer. We have a high-speed channel and a high power server.
The constant of our service demand allows us to keep low prices.
http://www.e-mailpromo.net
Please don't hesitate to contact us anytime!
We'll be happy to answer every your question!
http://www.e-mailpromo.net
We received your address from a public area.
We apologize if this letter has reached you by mistake.
We'll not disturb you any more.
NB. This message is sent in compliance with the new e-mail bill section
301. Under Bill S. 1618 TITLE III passed by the 105th US Congress.
This message can not be considered as Spam as long as we include the way
to be Removed, Paragraph (a)(c) of S. 1618.
TO REMOVE YOUR ADDRESS FROM YOUR MAILING LIST: you can send message to
info(a)e-mailpromo.net with Subject "Remove".