My July Reading List

Here we go again - no excuses - time to catch up on your reading:

• Baillie, R. T., G. Kapetanios, and F. Papailias, 2014. Modified information criteria and selection of long memory time series models. Computational Statistics and Data Analysis, 76, 116-131. 
• Pitarakis, J-Y., 2014. A joint test for structural stability and a unit root in autoregressions. Computational Statistics and Data Analysis, 76, 577-587.
• Ghysels, E., J. B. Hill, and K. Motegi, 2013. Testing for Granger causality with mixed data frequency. DP9655, Centre for Economic Policy Research.  
• Gresnigt, F., E. Kole, and P. H. Franses, 2014. Interpreting financial market crashes as earthquakes: A new early warning system for medium term crashes. Tinbergen Institute Discussion Paper TI 2014-067.  
• Marsh, P., 2013. A review of non-parametric econometrics. Econometrics Journal, 16, B1-B3(3). 

© 2014, David E. Giles

The Econometrics of Temporal Aggregation - I

Yesterday, I gave the A. W. H. Phillips Memorial Lecture at the 55^th Conference of the New Zealand Association of Economists. The conference was held at Auckland University of Technology in New Zealand, and it was a great success.

More on the conference itself in a subsequent post.

Meantime, you can find the slides for (a slightly extended version of) my address here, and an extended bibliography for the talk here.

In some posts that will follow I'll elaborate on some of the main messages that should be drawn from what I had to say at the conference.

© 2014, David E. Giles

The Econometrics Bard Strikes Again!

Guy Judge (now retired from the University of Portsmouth in the U.K.) has been back at it again in the past couple of days.

In recent years Guy has shared several original econometrics poems/songs with us through his Guy's Econometrics Blog.

The latest two contributions go under the titles of "Granger" and "Backing Bayes".

Enjoy them!

© 2014, David E. Giles

New Zealand Association of Economists Conference

In a couple of days' time I'll be heading off to New Zealand to participate in the 55th Annual Conference of the N.Z. Association of Economists. I'll be one of the keynote speakers, and I'm honoured to be presenting the A. W. H. Phillips Memorial Lecture.

That's "Bill" Phillips of The Phillips Curve fame - a very interesting and immensely talented New Zealander about whom I've posted previously, here and here.

My talk is titled, "The Econometrics of Temporal Aggregation: 1956-2014". The link to Bill Phillips is through his seminal work on continuous-time econometrics, and the lessons it has for econometric modelling when our data have been aggregated over time.

You can guess that I'll be posting on this topic in more detail in the near future. As soon as I've given my address, I'll make the slides available through this blog.

© 2014, David E. Giles

More on Celebrating Trygve Haavelmo

In a recent post I drew attention to the special issue of Econometric Theory that is being devoted to the contributions that Trygve Haavelmo made to econometrics, and to the founding of the Econometric Society. In fact, there will be two issues of the journal that will be dealing with this topic.

Most of the papers that will appear in the first issue (to be published next year) are now available on the ET website. One paper that isn't there yet is one that I mentioned in an earlier post. It's titled, "Trygve Haavelmo at the Cowles Commission", by Olav Bjerkholt. You can download this paper here.

Olav wrote to me recently, saying: "My own paper will appear in ET with some pictures and also an unusual illustration, a page from Haavelmo's notebook showing the list of persons who received his 1941 early version of Probability Approach." Olav is referring to Haavelmo's seminal paper, "The Probability Approach in Econometrics", which was published in Econometrica in 1944. That paper is available in its entirety, here.

Olav has kindly given me permission to reproduce Haavelmo's list, so here it is:

The First European Meeting of the Econometric Society

Olav Bjerkholt has alerted me to an interesting new paper of his that documents a milestone gathering of econometricians. Titled, The First European Econometric Society Meeting, September 1931, Lausanne, Olav's paper was presented at the 18th Annual ESHET Conference at the Université de Lausanne, last month.

You'll recall that I've mentioned Olav's work previously on this blog - here, here, and most recently, here.

Here's the abstract from Olav's paper:

"The idea of an econometric association was conceived in Europe in 1926, the organization meeting founding the Econometric Society (ES) took place in the U.S.A. in 1930, while the first ES meeting was convened in Lausanne at the end of September 1931. The venue was deliberately chosen to honour Walras and Pareto. The meeting was hastily prepared and had few participants. The Lausanne meeting established the tradition of Econometric Society European Meetings (ESEMs). The paper gives an account of the meeting with excerpts from the exchange between Council Members of ES in 1931. The participation, paper topics and the emphasis on paying homage to econometric pioneers at the Lausanne meeting is set out. The Econometric Society was the first international organization in economics. At the end of the first year ES had 163 members distributed over residents in 19 countries. The multi-language, multinational character of the original venture of bringing together scholars in Europe who shared an interest in the econometric program generated a series of ESEMs of considerable importance for the development of econometrics, until sombre political events overshadowed the meetings. The paper is part of a history project within the Econometric Society."

If you have an interest in the history of our discipline - and I think you should (!), then you'll find this paper extremely valuable.


© 2014, David E. Giles

An Extreme Publication Lag

We all complain about the delays associated with the academic publishing process. The referees can be very slow in reaching their recommendations; the revisions sometimes seem to be interminable; and then the accepted paper sits in a long queue awaiting its grand entry onto the world stage.

Occasionally - very occasionally - we encounter an extreme outlier in this process. I seem to recall that there was a paper by Paul Samuelson, written in the 1940's, that eventually appeared in print some decades later.

Olav Bjerkholt has kindly drawn my attention to an exceptional econometric example of delayed publication, involving an important paper by one of our founding fathers - Trygve Haavelmo. (Olav has written extensively and authoritatively on the early history of econometrics, and I've mentioned some of his contributions previously - here, and here.)

Now, what about this publishing delay?

Some Questions About ARDL Models

The majority of the blog-related comments and requests for help that I receive come from the one person - called "Anonymous". 

(S)he seems to have very broad interests.

Here's a very recent request for help relating to ARDL models - something that I've posted about here and here.

"I am working on income inequality. Can I use ARDL as I have only 27 annual observations? Also does ARDL itself takes care of problem of endogeneity? And what about if there is multicollinearity among explanatory variables - can we still use ARDL? Is any EViews code available to run ARDL?"

Taking the questions in order........

Do You Use P-Values and Confidence Intervals?

Unless your econometrics training has been true-blue Bayesian in nature, you'll have reported a lot of p-values, and constructed heaps of confidence intervals in your time.

Both of these concepts have been the centre of widespread controversy in the statistics literature since their inception. It's probably good to be aware of this - just so you don't go and "shoot yourself in the foot" at some stage.

Economist/econometrician Aris Spanos has published an interesting and readable piece about all of this In a recent issue of the journal, Ecology. His paper is titled, "Recurring Controversies About P Values and Confidence Intervals Revisited". You can read a summary on the Error Statistics blog, here.

I strongly recommend this paper.

© 2014, David E. Giles

New Award for David Hendry

It's difficult to imagine what our modern econometrics world would be like if it weren't for the numerous, seminal, contributions that Sir David Hendry has made over the course of his distinguished career.

So, it was wonderful to see this announcement two days ago from the Economic and Social Research Council:

"Professor Sir David Hendry has today received the ESRC Celebrating Impact Lifetime Achievement Award. Over five decades Professor Hendry has developed macroeconomic models capturing how economies work, which are now embedded in software widely used by policymakers and decision-makers."

You can read the full details of the award, and a related video, here.


© 2014, David E. Giles