Decline and Fall of Econometrics?

"Please bear in mind throughout that IT IS MEANT TO BE FUNNY."

(Evelyn Waugh, 'Author's Note' in Decline and Fall, 1^st ed., 1928)

I'm not sure that there's anything funny about the decline and fall of Econometrics, especially if you're an econometrician - that's if it were true, of course. But is it? Personally I don't think so.

Econometricians seem to be very much in demand. To take one example, if you take a look at the EconJobMarket site right now you'll see that over the past 6 months, and also in total since that site began, more jobs have been posted in the Econometrics category than in any other economics category (except for "Any Field").

BTW, don't be put off by the fact that there don't seem to have been very many jobs posted at all in the past 6 months - that period doesn't cover the annual recruitment season job posts. With that recruitment season  in mind, if you look back at last October's issue of JOE, you'll find that there were over 130 job postings in the "Mathematical & Quantitative Methods" (which is primarily Econometrics)category.

It will be interesting to see how the job numbers for econometricians stack up in JOE after this summer's break.

Imagine my alarm, then, when I obtained this chart while playing around the other day with Google Trends:

p-Values for Cointegration Tests With Breaks in the Data

In an earlier post I went through some econometrics that involved the problem of testing for multivariate cointegration in the case where there are one or more trend-breaks or level-breaks in the time-series data.  Specifically, I talked about the modified Trace tests introduced by Johansen et al. (2000), and I mentioned the really nice discussion of the application of these tests that is provided by Joyeux (2007).

Two things relating to this occurred to me recently. The first was that while I'd provided EViews code for calculating asymptotic critical values to be used with these tests, it would also be useful to have the corresponding code for calculating p-values for any calculated values of the Trace test statistics.

Second, given the discussion and comments in my recent posts (here and here) about open-source software, I thought it would be a good idea to make the p-values and critical values code available for users of R. (Thanks for the earlier comments, "Ben" and Tal Galili!)

So, in a joint effort, Ryan Godwin and I have written the R code, and extended the earlier EViews code to compute the p-values. Both of them are on the Code  page that goes with this blog - in two places: under this post, and also in place of the code for the earlier post. (You can thank Ryan for the nice windows that open when you run the R program.) In addition, an Excel workbook with a big selection of critical values is avalable on the Data page for this blog.

We hope you find the programs useful!


Note: The links to the following references will be helpful only if your computer's IP address gives you access to the electronic versions of the publications in question. That's why a written References section is provided.

References

Johansen, S., R. Mosconi and B. Nielsen (2000). Cointegration analysis in the presence of structural breaks in the deterministic trend. Econometrics Journal, 3, 216-249.

Joyeux, R. (2007). How to deal with structural breaks in practical cointegration analysis? In B. B. Rao (ed.), Cointegration for the Applied Economist, Second Edition, Palgrave Macmillan, New York, 195-221.

© 2011, David E. Giles

Your Very Own Theorem

Too many term papers to grade? Tenure clock running? Don't have the time to crank out a new theorem today?

Fear not! TheoryMine can come to your rescue. As they so delightfully explain:

"You can name your very own mathematical theorem, newly generated by one of the world's most advanced computerised theorem provers (a kind of robot mathematician), and you can immortalise your loved ones, teachers, friends and even yourself and your favourite pets".

And:

"You can buy new theorems which become yours to name. You will receive a printable certificate in PDF form of the theorem and its discovery (including an outline of the proof). You can then give this away, frame it, sing it, as you like!"

At only £15.00, this looks to me like a bargain just waiting to be snapped up! I mean, how long did you spend proving that last theorem of yours? I'll bet that (No. of Hours x Hourly Salary) > 15!

I particularly like the idea of "immortalising your teachers" (hint, hint), and I promise not to sing it!

HT to Ken Stewart.

© 2011, David E. Giles

gretl

In comments on a recent post, "Ben" and Tal Galili very sensibly asked if I could make R code available for the econometric analysis in my posts, in addition to EViews code. I'll be trying to do this wherever I can, given the time constraints.

The important point implicit in these comments is that R is free, open-source, software, whereas EViews is not. I'm definitely a supporter of open-source. Here's a suggestion that may be helpful in the meantime, especially if you aren't feeling up to learning R.

There's a nice open-source package called gretl that has much in common with EViews. It's specifically econometrics-oriented, with lots of the time-series features that are part of EViews' strength, and that are hard to match in a freindly way in a lot of other econometrics packages.

In case you're wondering, gretl is an acronym for Gnu Regression, Econometrics and Time-series Library.

The really good news is that is is very simple to open foreign data files in gretl, including EViews workfiles, SAS, STATA, and SPSS files. This might help some readers of this blog who don't have access to EViews.

Right now, there seem to be a few problems with opening some EViews 7 files in gretl - earlier versions of EViews are fine. Allin Cottrell is kindly checking this out, and I'll keep you posted on this point.

From here on I'll try and supply data in EViews.wf1, Excel, R, and STAT.dta files to maximize accessibility.

I'm only just starting to play around with gretl, but it looks just great!

(HT to Martina Lui - long overdue!)

© 2011, David E. Giles

An Important 'First'

There are lots of rewarding things about this job - the short hours, the long vacations every summer, the wonderful pay,.....................(yeah, right!)

But right up there among the very best things is when a grad. student gets their first publication acceptance! It's a really special day when that happens - and not just for the student.

Yesterday, I was thrilled that one of my grad. students, Ryan Godwin, achieved this important milestone, with a paper accepted for publication in Communications in Statistics - Theory & Methods.

One of the impressive things on this occasion is that the paper in question is not even part of Ryan's Ph.D. research. Extra frosting on the cake!

I'm sure that all of us can remember that first acceptance letter - or email, as it usually is these days. Here's to many more to come, Ryan!

© 2011, David E. Giles

Private Market for Personal Identity Data

Apparently, the British Government is planning to create a private data market for personal identity data. There's an interesting item abut this in Computer Weekly. 

You don't need to have a subscription to see the draft discussion document, "Identity Assurance (IDA) - Technical Infrastructure Services". Currently, it's accessible here.

One interesting thing that I learned is that "shared secrets" are apparently aka "memorable information" (see p.7 of the draft report).

Great timing, given the recent security breaches at Sony, Sega, and the like!

© 2011, David E. Giles

RStudio

If you're a user of the R statistical software environment (and you should be - if you're not, it's the best free lunch in town), then the (fairly) recently released RStudio user interface will undoubtedly interest you. Here's a quote from the description on their website, to give you the flavour:

• "RStudio brings together everything you need to be productive with R in a single, customizable environment. Its intuitive interface and powerful coding tools help you get work done faster.
• RStudio is available for all major platforms including Windows, Mac OS X, and Linux. It can even run alongside R on a server, enabling multiple users to access the RStudio IDE using a web browser.
• Like R, RStudio is available under a free software license that guarantees the freedom to share and change the software, and to make sure it remains free software for all its users. "

RStudio has been getting quite of a bit of attention - positive attention - especially in places that matter, such as R-Bloggers.

Anyway, I'd encourage you to check it out. If you're not already an R user, this interface might persuade you to take the leap. Just make sure that you update your installation of R to (at least) 2.11.1 before trying to use RStudio.

© 2011, David E. Giles

Invisible Econometrics

Welcome to The Museum of Non-Visible Art (MONA). Yes, you read that correctly. So what is this all about? Here's an extract from their site:

"The Non-Visible Museum is an extravaganza of imagination, a museum that reminds us that we live in two worlds: the physical world of sight and the non-visible world of thought. Composed entirely of ideas, the Non-Visible Museum redefines the concept of what is real. Although the artworks themselves are not visible, the descriptions open our eyes to a parallel world built of images and words."

You can buy (the ownership rights to) an imaginary piece of art that doesn't physically exist. For example, pledge $20 or more to the cause and you will receive:

On the Correlation Between Crime and Punishment

In two previous posts (here and here) I've taken a look at the OECD's 50^th anniversary present to all of us - their "Better Life Index" (BLI). One of the eleven components of that Index is titled "Safety" - and according to the OECD this component

"...... largely reflects the risks of people being physically assaulted or falling victim to other types of crime. Crime may lead to loss of life and property, as well as physical pain, post-traumatic stress and anxiety. The biggest impact of crime on people’s well-being appears to be through the feeling of vulnerability that it causes."

The BLI site also notes that

"Across the OECD, one person in six reported falling victim to a conventional crime, with physical assault accounting for nearly a quarter of all conventional crime. In terms of perceived safety, one out of four people in the OECD report feeling unsafe on the street after dark."

In addition, the average homicide rate across the OECD countries is 2.2 murders per 100,000 inhabitants.

An interesting question that arises is whether or not the relative safety level matches the relative incarceration rate.

Obesity, Driving, & Unbalanced Regressions

The Economist magazine's daily chart of 15 June 2011 was to do with a recent study on the relationship between obesity and amount of driving. This study is reported in a paper by Jacobson et al. (2011), which is "in press" at the journal, Transport Policy. This paper will bring tears to all econometric eyes! Not tears of joy, either.

There are so many things that one could say about it that it's really difficult to know where to start - but I'll try!

Before we start, though, a word about the charts in the 15 June chart blog in The Economist. There are two of them - look at the one on the left. Are they really plotting the correlation between the two variables in question? I don't think so.  Anyway, that's not my real gripe. My problem is with the paper that's being published by Transport Policy (TP). And judging by the nature of the comments on The Economist's blog, I'm not alone.

From its website, I see that TP has an impact factor of 1.024. This just goes to show how misleading impact factors can be - it's quite possible to get the impact factor above unity simply by publishing material that infuriates people so much that everyone has to cite it in order to take issue with it! Nice trick if you can get away with it!

Google Correlate

There's an interesting piece in the latest issue of The Royal Statistical Society's newsletter, RSSNews. It's about one of Google's new data analysis tools, called Google Correlate. If you haven't come across this, it's worth a look.

You can upload your own time-series data (from a .csv file, say) and then use Google's search history to find what other data your series correlates highly with.

Just keep in mind that "correlation" and "causation" are not the same thing! More on the latter in future posts.

© 2011, David E. Giles

Testing for OPEC-Causality

Last Friday, the OPEC meeting being held in Vienna broke up in what has been widely described as "disarray". A group of OPEC members, led by Iran and Venezuela, vetoed plans for an increase in the world production of oil. Think of the words "supply", "demand", and "elasticity". and you can guess what they were thinking.

Saudi Arabia then announced that it would step up to the plate and increase its production to meet world demand. It'll no doubt be assisted by some of its Gulf allies.

OPEC was founded in September 1960, with 5 member countries. Today, this cartel has 12 members - Algeria, Angola, Ecuador, Iran, Iraq, Kuwait, Libya, Nigeria, Qatar, Saudi Arabia, United Arab Emirates, and Venezuela. Yes, make no mistake about it - OPEC is literally a cartel. It determines the supply and price of crude oil. Let's take a closer look.

Type I Errors in Economics Publishing

"Economists are peculiar social scientists not least because they attach enormous value to the publication of articles in the refereed journals and virtually no value to the publication of books. It is difficult for economists to have a coffee break without a conversation which quickly turns to questions like: "Why was my article refereed by so-and-so journal? Why did the anonymous referees say what they did? Where shall I send my next paper?" In short, the publication process merits a hideous fascination if only because it governs the pecuniary and non-pecuniary rewards of the economics profession."

Sound familiar?

This quote is taken from the blurb (product description) for the book, Publishing Economics: Analyses of the Academic Journal Market in Economics, edited about a decade ago by Joshua Gans. This book contains fifteen great articles that still highly relevant, and should be required reading for all young econometricians.

Joshua, and his former Stanford class-mate, George Shepherd, wrote a great article (Gans and Shepherd, 1994) providing us with tales of rejection from many eminent economists. Econometricians can take heart from the fact that the list includes James Tobin's unsuccessful attempts to get his work on the Tobit model published in Journal of the American Statistical Association (the statisticians' equivalent to the AER.)

The material gathered in preparing this paper was subsequently edited by George into the book, Rejected: Leading Economists Ponder the Publication Process. Again, an except from the product description does a better job than I could:

"Many people know that Robert Merton and Myron Scholes won the 1997 Nobel Prize in Economics for their now famous 1973 paper on options pricing. A lucky few are aware that this options paper was rejected by several leading journals before the Journal of Political Economy finally accepted it. This book compiles tales of rejection from more than 70 leading economists, including 20 Nobel winners."

The tales herein are a great read. Especially if you've recently received a rejection letter!


Note: The link to the following reference will be helpful only if your computer's IP address gives you access to the electronic versions of the publications in question. That's why a written Reference section is provided.

Reference

Gans, J. S. and G. B. Shepherd (1994). How are the mighty fallen: Rejected classic articles by leading economists. Journal of Economic Perspectives, 8, 165-179.

© 2011, David E. Giles

Feeling Rejected?

Earlier this week I had a request to become a "contact" for a particular person on LinkedIn. As this didn't appear to be someone I knew, I was about to press the "Delete" button when I had this strange feeling that the name was just slightly familiar, after all. And then I remembered.

As Editor, or Associate Editor, of an academic journal, you occasionally get to see the darnedest things - and I don't just mean the things that are submitted! I'm referring to the dark underbelly of academia. That place where your supervisor should have warned you never to go. I have a number of editorial responsibilities right now, including being joint Editor of The Journal of International Trade & Economic Development. This year alone I've been saved on two occasions from making a bad mistake (and possibly getting our publisher involved in a lawsuit), only as a result of eagle-eyed referees. Hopefully, there aren't even more cases that I'm unaware of!

An Even Better Life?

Last week, in a post titled, It's a Wonderful Life, I discussed the OECD's recently released "Better Life Index" (BLI). The index is made up of eleven different components that are then aggregated, with equal weights, into a single BLI, which you can see here. In addition, the OECD provides an interactive tool so you can select your own weights across the 11 measures, and then make cross-country comparisons.

At the end of that earlier post I noted that one question that I hadn't addressed  was: "What's really driving the variability in the country rankings?" I promised to return to this question - so here we are.

Busking for Business

On World Statistics Day, 20.10.2010, the Royal Statistical Society launched its getstats campaign. This is a major drive to promote the statistical profession and the importance of statistics to the general population. It's a consciousness-raising venture that, as a Fellow of the RSS, I heartily endorse. Recently, the getstats campaign manager, Debra Hurcomb, has suggested that we should view this as a form of busking.

Now, taken literally, this appealed to me no end, and I thought immediately of the International Society for Bayesian Analysis (ISBA)  and their biennial meetings, which are famous for their highly creative cabarets. If you're not familiar with all of this already, you really should check it out at The Bayesian Songbook.

The Songbook should be on the "required reading" list of courses in statistical or econometric methodology. I certainly link to it on the web page I use when I'm teaching our elective graduate econometrics course - a course that includes both frequentist and Bayesian material. We don't actually have sing-along sessions in that course - given the quality of my singing, that would be one sure way to clear the room in a hurry.

The origins of the Bayesian Songbook can be traced to to George Box's wonderful song, "There's no Theorem Like Bayes' Theorem", to be sung to the tune of "There's no Business Like Show Business". It's a classic of epic proportions, and was first performed at the first ISBA Valencia meeting in 1979.

Just another instance of  Bayesians being ahead of the curve! As if we really needed more examples!

So, following the exhortations of the RSS, and the lead of the ISBA, it's just glaringly obvious that we should be out there promoting Econometrics with some creative and energetic busking. And we have a leader already! Guy Judge at the University of Portsmouth in the U.K. has been on to this for some time, so it's just a matter of getting in behind him! I'm particularly partial to his "Heteroskedasticity Blues"; and if (like me) you can't carry a tune, Guy also has a nice collection of Econometrics poems.

So, once more, from the top - with feeling!

© 2011, David E. Giles

Eliminating Inflation Through Creative Econometrics

The other evening my wife asked me if tulips are poisonous to cats. I didn't know the answer off hand. The question was prompted by the fact that our furry beast was attempting to graze in the remaining tulips on our patio. (Technically, it was attempting to browse, but that's O.K..) The same tulips I've been staring at for about 7 weeks now!

Here on the We(s)t Coast Spring has been longer, cooler, and more damp this year than any of us would care for. After a while it gets you down, and you feel as if the fog has actually penetrated your brain. Strange thoughts begin to surface, just when you least expect them.

The question about the cat and the tulips interrupted my own contemplation of a notice that I'd seen on campus earlier in the day. It concerned a summer course being offered by our Department of Linguistics here at UVic. namely LING 388: "An Introduction to Grammar of English Usage". I had been thinking that this was something I should recommend to some of our students - until I observed that it said at the bottom of the notice: "No Prerequisites Required." Isn't that third word redundant?

I warned you - this weather does bad things to your mind!

Zanran

Zanran is to data, what Google is to text.

Now in its Beta-testing phase, the new search engine, Zanran, seems certain to appeal to all of us who use data on a regular basis. Founders Jon Goldhill and Yves Dassas describe Zanran in the following way:

"Zanran helps you to find ‘semi-structured’ data on the web. This is the numerical data that people have presented as graphs and tables and charts. For example, the data could be a graph in a PDF report, or a table in an Excel spreadsheet, or a barchart shown as an image in an HTML page. This huge amount of information can be difficult to find using conventional search engines, which are focused primarily on finding text rather than graphs, tables and bar charts."

On 31 May I used Zanran to search "zero inflated Poisson" (see above), and returned 4,394 results, each linking to data in graphs and/or tables. Here are the first 5 items in the list:

When you "hover over" the pdf icons for each of the first three items above, here are the previews that you see:

Then, of course, if you like what you see, you just click your way to the linked document, be it a pdf file, an Excel worksheet, an HTML page, or whatever.

No, I'm not an investor, but this is one cool search engine. Try it - you'll love it!

(HT to Andrew Gelman.)

© 2011, David E. Giles

World Bank e-Atlas

The World Bank has recently released an "e-Atlas" that provides an interesting collection of maps and graphs for a wide range of economic development indicators. Over 170 indicators are provided for more than 200 countries.

Let's suppose that you want to graphically compare per capita GDP in PPP international dollars for Luxembourg, Norway, and the entire world, from 1980 to 2009. No problem! The data can be compared "on the fly" or exported for further analysis.

What about a comparison of the percentage of births attended by skilled health staff, in Haiti, Kenya, Somalia and Pakistan? It's just a click away. Did you know that this percentage declined from 50% in 1989 to 41.6% in 2003, in Kenya? It was still only 26.1% in Haiti in 2006.

You can also create "zoomable" maps that display the global "distribution" of the variables you're interested in. These maps have nice "hover over" features and can be animated to depict changes over time.

Here are a couple of examples to whet your appetite.

1. Maternal Mortality Rate (The Whole Wide World):

2. Mobile 'Phone Usage (Zoomed in to Selected Region):

This is a nice utility that's definitely worth checking out. And you even get to choose the colours!

© 2011, David E. Giles

Well, Knock me Down and Call me Susan!

For your edification I offer (without comment):

Brigitte H. Bechtold, "The Practice of Econometrics: A Feminist Critique", Review of Radical Political Economics, 1999, 31, 40-52.

Check it out, and enjoy (?) !

(HT to The Unbroken Window, 2008.)

© 2011, David E. Giles