Bookmark and Share

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]

st: Happy Holidays


From   Robert A Yaffee <bob.yaffee@nyu.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Happy Holidays
Date   Sat, 25 Dec 2010 01:11:02 -0500

Dear friends and family,                                                              Friday, December 24, 2010
         There are a group of people who each year marvel at the acceleration with which time seems to have passed.   Each year it flies by faster than before.   It was only a year ago, last January that we were holding our forecasting course at Baruch College.     Of course, the research on forecasting was compressed by our analysis of the Chornobyl survivors.    We had prepared a presentation of a poster for the National Science Foundation’s conference for grantees.   But Washington was inundated with snowfall, causing the closing of Reagan National Airport and the temporary paralysis of our capital.
         Our dataset was too large for the ordinary 32 bit computer.   We had more than 2000 variables, and our pilot data alone was enough to slow down the processing to a crawl and cause the program to drop the value labels and variable names to conserve enough space to process the simplest descriptive tasks.   I had to take the quantum leap from the common computer world to that of the 64 bit universe, with its capabilities for up to 194GB of RAM, along with a variety of incompatibilities between the 32 bit and 64 bit versions of software.
         The Dell 64 bit computer had plenty of RAM but it kept crashing.   In fact, it continued to crash for one reason after another eight times within ten months.  For some time I thought I could repair the crashed computer, but it became evident that the Dell computer no longer had the quality assurance they used to have.   Outsourcing the fabrication of parts may have been a cheaper short term solution, but the reliability of the parts suffered.   A  New York Times article quoted an employee of the National Security Agency of reporting that Dell was being sued for covering up defects in the computers they built and distributed and that Michael Dell was being sued personally as well.   At that point, I realized that I was wasting my time tilting windmills if I were to try again to repair the Dell M17X system myself.  I abandoned the machine and got a Toshiba 64 bit laptop with eight GB of RAM, which proved much more reliable and stable.   
	But time was flying by and we had to attend the OxMetrics Users Group down in Washington, DC..   The travel schedule was going to be a busy one.   I was invited to attend the OxMetrics conference in Washington, D.C.   After that,  we  were off to the International Symposium of Forecasting in San Diego in June.   Just after returning from that, I received an invitation from the Royal Scottish Institute for Research.  In two weeks, there would be a conference in honor of Prof. David Hendry’s being knighted by Queen Elizabeth last year.   I had sent Sir David a note of congratulations and perhaps that was how I was invited to the conference at the University of St. Andrews for the commemorative conference.   When I got to St. Andrews,   I found Rob Engle,  our Nobel Laureate in economics was there,  along with Neil Ericsson from the U.S. Federal Reserve Board,  along with Drs. Katrina Juselius,  Soren Johansen,  Siem Jan Koopman,  and Aris Spanos,  among many other well-known !
 econometricians from around the globe.  Prof.  Halbert White came over from UCSD.   He regaled playing swing and popular songs on the trumpet with the band at the reception for Sir David at the St. Andrews Golf Course Club house restaurant.   The wine was out-of-this world and the sumptuous food was had by all.   
	I returned to the U.S. so Rosemarie Foster and I could prepare the analysis for our poster at the rescheduled NSF conference for grantees to be held after I returned from London.   But I also had to prepare my presentation on forecast evaluation with Stata software at the London Stata Conference.    About three days after returning from London, Rosemarie and I were off to Arlington for the NSF conference. By December 1st, I had to return to London to attend the High Dimensional Econometrics Conference at the Cass Business School and to teach my forecasting course.   No sooner had I arrived than the blizzard came which paralyzed London. Two of my students could not make it to my class.  But  we persevered. After four days of teaching for eight hours/day,  I went up to Oxford for a memorial service for Ana Timberlake who had passed away a year ago around Christmas time.   The ceremony was wonderful and I heard that some of my friends had gotten a grant from George Soros to co!
 ntinue with their research.  I am delighted for them.   It was getting late and the hosts had arranged for a cab to take some of us back to London.   But the cab was so hot, I lowered the window to cool off.   In the process, I caught a good chill which triggered my asthmatic cough.  So, instead of trying to attend another conference, I spent the weekend in bed, coughing up a storm.    I brought green tea, Hall smoothers, and Ricola lemon mint cough drops with me and on the plane ordered hot water in which to homebrew my cough suppressant.  
My asthma specialist was so alarmed at my condition, he treated me to an alburterol sulfate nebulizing which gave me the first real relief I experienced since I contracted the cough.  He also gave me a prescription for a nebulizer which along with some other medicine, has helped me recover. But today, Christmas Eve is the first day I have really had my voice back.   I haven’t had time to do much in the way of buying gifts or sending out cards.   Right now,   I have to prepare some Bayesian programs to perform space –time analysis of the symptomatological patterns of the Chornobyl survivors once our data is all collected.    There is always much to do. 
There have been rumors that the Belgians and others are hosting tours to the hot zone around Chornobyl.   It’s astounding how some do not seem to not understand the hazards of ionizing radiation and what happened to those who were sent in to clean up the radiation.  Most are no longer with us.  The survivors have plenty to complain about.        
If you haven’t seen it, two movies are de rigeur:   Inside Job by Charles Ferguson and Fair Game with Sean Penn and Naomi Watts.  I highly recommend both of them.
Meanwhile I continue working on my books.   But I will reserve comment on those till later.    For the time being, let me with you and yours a Happy Holiday and wonderful New Year.

    Bob Yaffee


Robert A. Yaffee, Ph.D.
Research Professor
Silver School of Social Work
New York University

Biosketch: http://homepages.nyu.edu/~ray1/Biosketch2009.pdf

CV:  http://homepages.nyu.edu/~ray1/vita.pdf

----- Original Message -----
From: David Greenberg <dg4@nyu.edu>
Date: Wednesday, December 22, 2010 6:05 pm
Subject: Re: st: Dispersion parameter for a Negative Binomial model within GEE framework
To: statalist@hsphsun2.harvard.edu


> The negative binomial regression model is not a fix for zero 
> inflation, only for over-dispersion. David Greenberg, Sociology 
> Department, New York University
> 
> ----- Original Message -----
> From: a b <andythezoologist@hotmail.com>
> Date: Wednesday, December 22, 2010 4:21 am
> Subject: st: Dispersion parameter for a Negative Binomial model within 
> GEE framework
> To: statalist@hsphsun2.harvard.edu
> 
> 
> >  Dear Statalisters,
> > 
> > 
> >  I have repeated measures data and I want to model at the population 
> 
> > average level, and hence I am using GEE. My data is also count data 
> 
> > and quite zero inflated - so I am modeling with a negative binomial 
> distribution.
> >  I was wondering how best to estimate the dispersion parameter 
> (alpha 
> > in Stata/k in Hardin and Hilbe) for the model?
> >  Do I just run a nbreg model, ignoring the repeated nature of the 
> > data, and take the alpha estimate from there and subsitute it into 
> the 
> > GEE model?
> > 
> >  Many thanks for any help!
> > 
> >                                                
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/help.cgi?search
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index