Thursday, January 31, 2013

How to Build a Great Model -- Insights from Hal Varian

I have chaired 16 doctoral dissertation committees and tomorrow my 17th doctoral student will be defending his dissertation. In academia, we sometimes refer to the chair of a student's doctoral dissertation committee as his/her "advisor."

So, necessarily, a conscientious advisor offers advice to students during his/her doctoral studies especially when it comes time to doing research for the dissertation. Here one may help in selecting the topic, suggesting related problems, assisting a student in getting papers out for publication in journals, listening to a student give a talk/seminar, etc.

It is rewarding, as a faculty member, to see a student's progress and the greatest moment comes after the dissertation is successfully defended and the student receives the diploma wearing the cap and gown. Having a student get a really nice job is also important (to the student and his family and also to the advisor who has also invested a lot of time).

I regularly offer my students advice -- even what publications they should read and send them links to interesting articles -- these can be in newspapers or magazines or journals or even working papers.

Of course, it also is very helpful to read articles on how to do great research and to write papers.

Hal R. Varian who needs no introduction to many and is now the Chief Economist of Google and an emeritus professor at UC Berkeley, having been the founding Dean of the School of Information there, wrote a piece, "How to Build an Economic Model in Your Spare Time," which is very useful not only to economists, but to operations researchers, and mathematical modelers, in general. An aside, my first book, "Network Economics: A Variational Inequality Approach," appeared  in 1993 was the first volume in the Advances in Computational Economics series by Kluwer (now Springer), and then appeared in revised form in 1999. And, Hal Varian was on the editorial board as was Dan McFadden, who has since received the Nobel Prize. Plus, Hal will be giving a plenary talk at EURO -- INFORMS, in Rome, Italy, July 1-4, 2013.; the link to the conference website is here.

His advice:

  • Look for ideas in the world, not in the journals. (especially The New York Times, The Wall Street Journal, and The Economist -- all personal favorites of mine, as well)
  • First make your model as simple as possible, then generalize it and you may go back and forth. (Try out simple examples.)
  • Look at the literature later, not sooner.
  • Model your paper after your seminar. (Here he emphasizes the importance of presenting your work to an audience. He also suggests setting your paper aside for awhile and then going back to it.)
  • Stop when you've made your point.
I loved how he compared developing a model and the associated creative process to sculpting. I take great pleasure in my research and do get a thrill like when I paint or look at art when the model makes sense, is elegant, and also abstracts something relevant and interesting in the real world.

According to Hal Varian: This back-and-forth iteration in building a model is like sculpting: you are chipping away a little bit here, and a little bit there, hoping to find what's really inside that stubborn block of marble. I choose the analogy with sculpting purposely: like sculpture most of the work in building a model doesn't consist of adding things, it consists of subtracting them. This is the most fun part of modeling, and it can be very exciting when the form of the idea really begins to take shape. I normally walk around in a bit of a daze at this stage; and I try not to get too far away from a yellow pad. Eventually, if you're lucky, the inner workings of your model will reveal itself: you'll see the simple core of what's going on and you'll also understand how general the phenomenon really is.