James Buchanan

James M. Buchanan, Nobel Laureate in 1986 for his work in public choice, has passed away at the age of 93. The New York Times reports on his death here.

I recall Buchanan presenting a seminar in the Department of Economics at the University of Canterbury (New Zealand) around 1990 or 1991. It was a great seminar, but what stands out most for me is the question period that followed his presentation.

A question was posed, which Buchanan answered in a clear and convincing way. An M.A. student then raised his hand and commented that he thought that the answer to the question should really be  quite different to what Buchanan had said. After thinking for a moment, Buchanan smiled, and said: "You know, I think your answer is much better than the one that I gave".

Everyone in attendance was mightily impressed - by our speaker's humility.

© 2012, David E. Giles

EViews 8

A Beta version of EViews 8 is available for users of version 7 of the package. You can get more information, and download the Beta version, here.

I've downloaded the 64-bit version of EViews 8, and I've started to play around with it. The advantage of the 64-bit version is that it can handle 120 million observations in a page, whereas the 32-bit version is limited to about 15 million observations. There are similar increases in the number of objects per workfile that can be handled. Of course, your machine must have a 64-bit OS, and lots of RAM to achieve these gains.

What else is new in EViews 8?

The list is long, but it includes the following:

• Improvements to the general interface; data handling; and graphing.
• Census X-13 for seasonal adjustment.
• Panel principal components analysis.
• Panel cointegration estimation.
• Panel causality testing.
• Heckman selection models.
• Bayesian VARs.
• Markov switching models.

And there's lots, lots more.

If you're an EViews user, take a look and see what interests you.

© 2012, David E. Giles

Good Prospects for Econometrics Grad. Students?

Hat-tip to Ben Atkinson for drawing my attention to the fact that in The Chronicle of Higher Education on 17 December, Sara Hebel had a piece titled, "Academic Job Openings for Economists Grew Again This Year".

Here's what she had to say:

SPEED Dating at the JSM

The Joint Statistical Meeting (JSM) is held each summer in a city in the U.S.A. or Canada. Last year, the JSM was held in San Diego, and it attracted some 6,300 attendees. This August, the JSM will be held in Montréal.

In an effort to deal with the size of the meeting, and the number of parallel sessions, a new format for some "contributed" sessions is to be tried out this year.

The following is an extract from the latest issue of Amstat News, the newsletter of the American Statistical Association.

On Random Numbers

These days, we take it for granted that it's very simple, and relatively costless, to generate random numbers. Really, these are "pseudo-random numbers", and we should pay careful attention to the quality of the algorithms that are used to generate them, as I've noted previously (and also here) on this blog.

Putting these qualification to one side, however, we've certainly come a long way from having to rely on that famous tome, A Million Random Digits With 100,000 Normal Deviates. Published by the Rand Corporation in 1955, this classic provided (lots of!) random digits that could be used as an aid to simple random sampling, as well random deviates generated from the standard normal distribution. The latter could be used in Monte Carlo simulations, for example.

The following is an extract from a handout that I use in my introductory statistical inference course:

Sad News

I was sad to learn that John Nankervis passed away on Christmas morning. Well known for his many important contributions to econometrics, John was Professor of Finance and Director of Research in the University of Essex Business School.

John will be greatly missed.

© 2012, David E. Giles

The Greenfield Challenge

The European Network for Business and Industrial Statistics has the objective of "...connecting individuals and organisations, interested in theoretical developments and practical applications in the field of business and industrial statistics." They achieve this through their electronic network, and an annual Spring Conference.

The co-called "Greenfield Challenge", which is now an annual event, arose from a challenge laid down by Tony Greenfield in his George Box Medal acceptance speech in 2009:

"My challenge to you is that you will tell some audience about work you have done, and completed successfully because you used a statistical method. But that audience must be of people who are not statisticians.

And you will have spoken to those people through publications that are for the wider public, through magazines or newspapers, or from a public platform. You might even write a short story or a play.… That is my challenge: Tell the world, outside your circle, of work you have done, and done successfully because you used statistics."

Fighting words!

You can read more about the responses to this challenge from members of the ENBIS community here.



I really like this challenge. It seems to me that it would be equally applicable to the econometrics community.

What do you think?

© 2012, David E. Giles

National Econometricians' Day

The Netherlands has been a power-house of econometrics since day one. Think of Jan Tinbergen's contributions to the birth of the discipline, for instance. 

So, perhaps it will come as no surprise that 5 February 2013 will be National Econometricians' Day, or "LED" (Landelijke Econometristendag), for students of econometrics. Billed as ".... the largest annual career event for econometricians in the Netherlands", the one-day event will be held in Utrecht in 2013. Their slogan - "LED it be your day!"

I've blogged previously (here and here) about another Netherlands-based initiative for students of econometrics - the Econometric Game. This event has gone from strength to strength, and attracts teams of students from all over the world.

It's great to see the extent of organization and involvement among the Dutch students. They have six associations for econometrics students, and these form the Landelijke Orgaan der Econometrische Studieverenigingen (LOES). The individual associations are: Asset | Econometrics (Tilburg University), Econometrisch Dispuut (Erasmus University), Kraket (VU University), SCOPE | Vectum (Maastricht University), VESTING (University of Groningen) and VSAE (University of Amsterdam).

© 2012, David E. Giles

Top Posts for 2012

Here are the "Top Ten" new posts on this blog during 2012. They're ranked in order of popularity (1 = top)  based on page-views:

1. My "Must Read" List (26 September 2012)
2. An Overview of VAR Modelling (23 March 2012)
3. Regression and Causation (4 December 2012)
4. Seminar Attendance: A Pep-Talk for Grad. Students (8 October 2012)
5. Modelling Extremes (16 April 2012)
6. Integrated & Cointegrated Data (5 June 2012)
7. Degrees of Freedom in Regression (12 October 2012)
8. Central Limit Theorems (29 October 2012)
9. Listening to Your Data (31 October 2012)
10. What I Learned Last Week (12 October 2012)

Thanks for your participation and your comments!

© 2012, David E. Giles

Eggnog With an Econometrics Flavour

With Christmas almost upon us I was giving some thought to a seasonally-themed post. The truth is, it's not as easy to produce one with an econometric flavour as you might think.

I suppose we could look at forecasting retail sales for the holiday season; or construct a SUR model for the market shares held by the major courier companies as they deal with all of us last-minute shoppers and givers. But I wanted something that clearly boasts "Christmas" in the title.

Interestingly (or not) there are not that many published applied econometrics papers that fit the bill. A few years ago, however, I struggled to fill this void with a paper I titled, "Testing for a Santa Claus Effect in Growth Cycles". (You can download the working paper version from here.)

The abstract tells the story:

"We examine the seasonal distribution of turning points in the post-war growth cycles of sixteen economies. Using nonparametric tests for distributions on the circle, we cannot reject a uniform distribution for the turning points for most of the countries. In the case of troughs, uniformity is supported for eleven countries, notwithstanding the unusually large number of December or January troughs. This provides evidence against a ‘Santa Claus effect’ in the growth cycle."

This was one of the first papers that I wrote that involved goodness-of-fit testing with "circular" data. You'll find other posts on this general topic here and here, if you're interested. There are a few things I'd do differently if I were writing the Santa Claus paper now - but isn't that always the case?


Reference

Giles, D. E., 2005. Testing for a Santa Claus effect in growth cycles. Economics Letters, 87, 421-426.

© 2012, David E. Giles