Dear all,
I am working with interactive variables in Clarify and I would like to know
if I can create macros in order to simulate quantities of interest without
setting values by hand.
My model contains three continuous by continuous interaction terms (X1X2,
X1X3 and X1X4). The first variable, X1, has an interval that goes from 0 up
to 10, the other three variables go from 1 up to 5. Is there a way to make
simulations with all the possible combinations of values for the three
interactions by using macros? If so, would you have any sugestions for such
a macro?
This is the first time that I work with Clarify, so please excuse me if
this is a basic question. Any help would be greatly appreciated.
Best regards,
Alessandro Freire
Dear Clarify users,
I have a question on a matter I have already touched upon in a previous message (October 2011). I didn't receive any answer so I am asking again trying to be more clear.
My point is about the calculation of expected values from a linear regression of a logged variable. -Simqi- with the option -tfunc(exp)- should yeald results transformed back to the original (unlogged) scale. Stata's official command -adjust- with the option -exp- should do the same. However the two do not coincide. What Stata appears to do is the simple exponentiation of the expected value: exp[E(lny)]. This yealds the expected geometric mean of the unlogged variable (geometric mean is always less than arithmetic mean and more similar to the median for skewed variables). -Simqi, tfunc(exp)- yealds results flagged E[exp(lny)]. Actually the latter always display higher values than the former, as if they were more similar to expected arithmetic mean. So my question is: how does -simqi- exactly compute expected values when I ask for the -tfunc(exp)- option?
Many thanks
Renzo Carriero
Dear Clarify users,
I have a question about the tfunc(exp) option of simqi command. I can't understand how this option exactly compute exponentiated expected value. My
depvar is a log ratio, say y=ln(a/b). My linear regression model contains a 4-category nominal variable (indicated by 3 dummies) plus covariates. When I
compute mean predicted values on the log scale, results yielded by simqi coincide with those yielded by Stata's command adjust. On the contrary, when
I ask for expected values in the original scale, that is exp(y), they markedly differ. It seems that Stata (adjust command with the "exp" option)
exponentiates the mean expected value, such as exp[E(y|x)], while simqi does not. So what is exactly the way in which simqi transforms expected values in
the original scale?
To help, I add below a simplified example of what I mean using Stata's auto dataset
Many thanks
Renzo
--
Renzo Carriero
Dipartimento di Scienze Sociali
via S. Ottavio 50
10124 Torino - Italy
+390116702658 (office)
+393898160069 (mobile)
+390116702612 (fax)
sysuse auto.dta
g lny=ln(price/mpg)/*generate a log var similar to mine*/
xtile x=length, nquantile(4) /*generate a 4 category variable from length variable"
tab x, gen(x)/*generate 4 dummies*/
reg lny x2-x4/*x1 is omitted as reference category*/
Source | SS df MS Number of obs = 74
-------------+------------------------------ F( 3, 70) = 18,48
Model | 10,7554461 3 3,58514872 Prob> F = 0,0000
Residual | 13,580584 70 ,194008343 R-squared = 0,4420
-------------+------------------------------ Adj R-squared = 0,4180
Total | 24,3360302 73 ,333370277 Root MSE = ,44046
------------------------------------------------------------------------------
lny | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
x2 | ,576453 ,1461643 3,94 0,000 ,2849374 ,8679685
x3 | ,635685 ,1376187 4,62 0,000 ,3612132 ,9101569
x4 | 1,050355 ,1437037 7,31 0,000 ,7637466 1,336962
_cons | 5,078354 ,0961171 52,84 0,000 4,886655 5,270054
------------------------------------------------------------------------------
adjust x2=0 x3=0 x4=0 /*compute expected value for omitted category 1*/
Dependent variable: lny Command: regress
Covariates set to value: x2 = 0, x3 = 0, x4 = 0
------------------------------------------------------------------------------------------------------
----------------------
All | xb
----------+-----------
| 5,07835
----------------------
Key: xb = Linear Prediction
adjust x2=0 x3=0 x4=0 , exp /*compute expected value for category 1 in the original scale (price/mpg)*/
Dependent variable: lny Command: regress
Covariates set to value: x2 = 0, x3 = 0, x4 = 0
------------------------------------------------------------------------------------------------------
----------------------
All | exp(xb)
----------+-----------
| 160,51
----------------------
Key: exp(xb) = exp(xb)
estsimp reg lny x2-x4/*replicate analysis with estsimp*/
setx 0 /*set x1=1*/
simqi
*this output omitted, it is roughly equal to that of adjust without exp option
simqi, tfunc(exp)
Quantity of Interest | Mean Std. Err. [95% Conf. Interval]
---------------------------+--------------------------------------------------
E[exp(lny)] | 177,6523 17,95269 145,2497 215,657
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=clarify
There was a migration issue with clarify while upgrading
"gking.harvard.edu", the files have now been restored and will work as
expected.
Thanks,
-Richard Brandon
rbrandon at iq.harvard.edu
------------------------------------------------------------------------
Hello,
I'm having trouble installing Clarify. I have Stata10/IC installed on
a Windows 7. When I entered the command: "net from
http://gking.harvard.edu/clarify/" <http://gking.harvard.edu/clarify/%22>;, I get the following error message:
filehttp://gking.harvard.edu/clarify/stata.toc not found
http://gking.harvard.edu/clarify/ either
1) is not a valid URL, or
2) could not be contacted, or
3) is not a Stata download site (has no stata.toc file).
r(601);
I tried downloading the .zip and attempting the comparable command
(for the local directory) but had the same problem. Any advice?
Thanks!
-Thomas
--
Thomas J. Leeper
PhD Candidate
Coordinator, Undergraduate Research Participation Pool
Department of Political Science
Northwestern University
leeper@xxxxxxxxxxxxxxxxxx
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:http://lists.gking.harvard.edu/?info=clarify
Hello,
I'm having trouble installing Clarify. I have Stata10/IC installed on
a Windows 7. When I entered the command: "net from
http://gking.harvard.edu/clarify/", I get the following error message:
file http://gking.harvard.edu/clarify/stata.toc not found
http://gking.harvard.edu/clarify/ either
1) is not a valid URL, or
2) could not be contacted, or
3) is not a Stata download site (has no stata.toc file).
r(601);
I tried downloading the .zip and attempting the comparable command
(for the local directory) but had the same problem. Any advice?
Thanks!
-Thomas
--
Thomas J. Leeper
PhD Candidate
Coordinator, Undergraduate Research Participation Pool
Department of Political Science
Northwestern University
leeper(a)u.northwestern.edu
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=clarify
On your webpage:
http://gking.harvard.edu/clarify/
and:
http://gking.harvard.edu/publications/clarify-software-interpreting-and-pre…
it states:
To install Clarify, type "net from http://gking.harvard.edu/clarify" at the Stata command line.
This does not work. Stata gives this error message:
file http://gking.harvard.edu/clarify/stata.toc not found
http://gking.harvard.edu/clarify/ either
1) is not a valid URL, or
2) could not be contacted, or
3) is not a Stata download site (has no stata.toc file).
r(601);
Installing Stata programs from zipfiles is a fairly painful
experience for average Stata users since they tend not to know
which directory ado and hlp files need to go in. So being able
to install Clarify via Stata's -net install- command would be
much better.
Hope this helps,
Dan Blanchette
Research Associate
Center for Entrepreneurship and Innovation
Duke University's Fuqua School of Business
dan.blanchette(a)duke.edu
---------- Forwarded message ----------
Date: Thu, 23 Dec 2010 12:07:37 -0500
From: Gary King <king(a)harvard.edu>
To: Dan Blanchette <dan.blanchette(a)duke.edu>
Subject: Re: [Scholars at Harvard] instructions to install -clarify-
Thanks for your note; can you send exactly what you did to the "Clarify Listserv" <clarify(a)lists.gking.harvard.edu<mailto:clarify@lists.gking.harvard.edu>>, and we'll figure it out.
Gary
--
Gary King - Albert J. Weatherhead III University Professor - Director, IQSS - Harvard University
GKing.Harvard.edu<http://gking.harvard.edu/> - King(a)Harvard.edu<mailto:King@Harvard.edu> - @kinggary<http://twitter.com/kinggary> - 617-500-7570 - Asst 495-9271 - Fax 812-8581
On Thu, Dec 23, 2010 at 12:04 PM, <Dan.Blanchette(a)duke.edu<mailto:Dan.Blanchette@duke.edu>> wrote:
Gary King,
Anonymous has sent you a message via your contact form (http://gking.harvard.edu/contact_owner) at Scholars at Harvard.
If you don't want to receive such e-mails, you can change your settings at http://gking.harvard.edu/cp/settings/site.
Message:
Hello Gary,
I just received an e-mail from a user of -intgph- (I am a co-author of it)...and I assume
you know -intgph- uses the -clarify- commands. The -intgph- user reported that he
was not so easily able to install -clarify-. I noticed on your webpage that you suggest
using the Stata command:
. net from http://gking.harvard.edu/clarify
This does not work.
I hope this feedback helps,
Dan Blanchette
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=clarify
Hi there,
I was wondering if you might be able to help me with a query.
I am using the STATA program CLARIFY (Michael Tomz, Jason Wittenberg, and Gary King (2001). CLARIFY: Software for Interpreting and Presenting Statistical Results. Version 2.0 Cambridge, MA: Harvard University, June 1. http://gking.harvard.edu)
I'm using it to get a set of predicted scores from an OLOGIT model I've been running.
So, I run the model. Then...
setx mean
setx div 0.1
simqi
...this gives me a set of predicted scores for each category of my OLOGIT dependent variable:
Quantity of Interest | Mean Std. Err. [95% Conf. Interval]
---------------------------+--------------------------------------------------
Pr(newsre~t=1) | .025927 .0027974 .0209661 .0317577
Pr(newsre~t=2) | .1168595 .0059575 .1055424 .128563
Pr(newsre~t=3) | .6739712 .0090134 .6563168 .6909184
Pr(newsre~t=4) | .1832423 .0074309 .1699783 .1987786
I want to create a graph in excel using the top two categories combined: so 0.674 + 0.183 = this gives me a predicted score of 0.86
Here's my question. I also want to add CONFIDENCE INTERVALS to this value using the C.I.s provided by CLARIFY. Do I simply add the C.I.'s together for the top two categories as I did with the predicted scores?
e.g. the C.I.s for this predicted score would be min. 0.656 + 0.169 and the max. would be 0.691 + 0.199?
So, the C.I.s of my predicted score of 0.86 (minimum 0.825) and (maximum 0.888)?
Is this correct?
Thanks in advance
James
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=clarify
Dear Clarify listers,
I have a model including indirect effects:
A. y1=f(x1, x2, y2, y3)
B. y2=f(x1, x2)
C. y3=f(x1, x2)
I am using the CLARIFY plugin in Stata to predict the effect of x1 on y1. Clarify is very useful to predict the outcome of each equation individually with the simqi command, but as far as I understand it this command does not assess the role of indirect effects.
Can Clarify make a prediction considering the individual direct effect of x1 on y1 as well as the indirect effect of x1 on y1 through y2 and y3? Can it also calculate the total effect of x1 on y1?
Any help in this regard would be greatly appreciated.
Best regards,
Thomas
McGill University
Thomas.Didier(a)mail.mcgill.ca<mailto:Thomas.Didier@mail.mcgill.ca>
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=clarify
I was recommended to this list to help solve a problem I'm having using
CLARIFY. I've got a dataset that includes an individual weight (wt_nation)
that allows the data to be considered nationally representative. So, when I
run ologit for the model I include [iw=wt_nation] at the end of the list of
commands. However, I don't seem to be able to get CLARIFY to work unless I
eliminate this part of the initial command.
I see from the help files that there seems to be a way to do this, but I'm
having trouble deciphering exactly what I need to do. If anyone has done
this before or has suggestions I would really appreciate hearing them.
thanks,
Melissa
*********************************************
Melissa R. Michelson
Visiting Fellow, Stanford Center for the Comparative Study of Race and
Ethnicity (CCSRE)
650.888.4810
Good afternoon, my name is Ryan Barkan and I am the owner of Creative Ventures, Inc. I am interested in finding out if you are in need of shirt printing for upcoming summer. We print hundreds of different garments including t-shirts, jerseys, sweatshirts and more.
You can visit our website at http://www.onehourtees.com.
I know the school year is just beginning and things may be a little hectic, but I wanted to contact you as soon as possible to help save you money and make the process of ordering garments a little easier this year. We currently provide over 175 schools in the Chicago-land area with clothing. We have done many jobs for Chicago Public Schools and work with many High schools in the Chicago suburbs. In addition we have worked with Northern Illinois Univ., Southern Illinois Univ., UCLA, University of Illinois Chicago, University of Illinois, and many more colleges. We also work with many YMCA organizations and Day camps throughout the area. I am emailing you today because our pricing is very competitive (which means we beat any price), we are conveniently located in Chicago and if you order over 30 shirts....
WE DELIVER NATIONWIDE FEDEX FOR FREE!
LICENSED VENDOR
Our Full Pricesheet is Available Online. Please Click Here to View it!
*Also Take Advantage of Our No Limit T-Shirt Special*
FOR ONLY $5.99 (5-7 Day Turnaround Time)
This includes any quantity of White, Ash, Heather and Natural Colored Tees with the front printed in FULL COLOR
Quicker Turnaround Time is Available. Please check our website for details.
No Screen Fees, No Setup Fees, No Design Fees, Super Quick Turnaround Time (As fast as 1 Hour)
1) We can print in full color unlike other Screen-printers
2) We Design Everything Free
3) We Deliver Almost Everything Free (over 25 shirts)
4) We NEVER charge you any setup or screen fees
5) We are able to meet critical deadlines if you need shirts quickly
This is strictly a courtesy email and if you are a person who could use our services, I would love to speak with you about starting up a long relationship that could benefit both of us. You can take a look at our website at http://www.onehourtees.com. If you would like references please let me know. Hope to hear from you soon.
Thank You,
Ryan Barkan - Owner
Creative Ventures, Inc.
OneHourTees.com
3056 N Oakley Ave 1N Chicago, IL 60618
Local: 1-773-687-0520
ryan(a)cvichicago.com
Click here to Unsubscribe
if you previously unsubscribed we apologize but our database was corrupt and we were unable to remove your address. We will make sure this doesn't happen again.