Bayesian Econometrics - Forty Years On

My Ph.D. dissertation was in Bayesian Econometrics. I started working on the dissertation early in 1973, and Arnold Zellner's classic text had been available for just over a year. So, perhaps not surprisingly, one of the first things that I did was to sit down and go through his book with a fine tooth comb. It took quite a while, but it was time well spent!

More on Confidence Bands for the HP Filter

Last December I had a post (and a subsequent correction) relating to a method for constructing confidence bands for the Hodrick-Prescott (H-P) filter. More specifically, I proposed a way of constructing a confidence band for the trend, or long-run growth component, of a time-series that has been run through the H-P filter.

A few people suggested to me that it might be worth writing up the material more formally, so I've done just that. You can find the working paper version here.

I've included two applications in the paper - both of them different from the one that I used in the blog post. One application relates to the U.S. unemployment rate, and the other involves U.S. real value-added output.

Feedback would be appreciated!

© 2012, David E. Giles

Extremes, the Generalized Pareto Distribution, and MLE

In a recent post I discussed some of my work relating to modelling extreme values in various economic data-sets. The work that my colleagues and I have been undertaking focuses on the use of the Generalized Pareto distribution (GPD). The estimation of the parameters of this model facilitates estimates of Value at Risk (VaR) and Expected Shortfall (ES).

There are various ways of estimating the parameters of the GPD but, not surprisingly, maximum likelihood estimation (MLE) is a common choice. However, there are some real traps when it comes to estimating the GPD using MLE, and they're worth knowing about if you're into this sort of thing.

Surplus-Lag Granger Causality Testing

My previous posts (here, here, and especially here) on Granger causality testing have attracted more interest than I anticipated. One of the things that I've discussed at some length is the "surplus-lag" approach that can be used when the data are possibly non-stationary and possibly cointegrated. In particular I've talked about the Toda and Yamamoto (1995) procedure, but there are alternatives such as those introduced by Dolado and Lütkepohl  (1996) and Saikkonen and Lütkepohl (1996).

These modifications to the standard approach to testing for Granger (non-) causality are needed to ensure that the Wald test statistic has its usual chi-square asymptotic null distribution. You can't just test in the usual way unless the data are stationary. In fact, the "surplus lag" approach has advantages even beyond those that we knew about already.

The Journal of Universal Rejection

Handling the econometrics submissions to this journal shouldn't be too onerous a task!

© 2012, David E. Giles

Modelling Extremes

Modelling extreme events is a challenging business. By definition, you're dealing with observations that are way out there in the tail(s) of the distribution. But that's where a lot of exciting things happen!

The Popularity of Statistical Packages

No matter what your favourite statistical package is, you'll find this post by Robert Muenchen highly informative.

Robert concludes that:

"By most of the measures discussed here, R is competing well with the commercial software vendors. However, I advise not over generalizing from this data. SAS and SPSS continue to dominate the corporate world and Stata is doing quite well in the scholarly arena. Each of these packages is dominant in one market or another."

© 2012, David E. Giles

Simultaneous Equations Models

Simultaneous Equations Models (SEM's) played an absolutely central role in the history of Econometrics. Simultaneous systems and measurement errors went hand in glove in forcing the emergence of econometrics as a field in its own right.

It's not that long ago that courses in econometrics devoted a good deal of time to SEM's. These days we spend much less time discussing these models, which is a shame because there are lots of important insights associated with them.

Count Data & the Hermite Distribution

One of the limitations of the usual discrete distributions that we use when modeling "count data" is that they can't allow for multi-modality (except in a trivial manner). So, there's no use in trying to model multi-modal data using a Poisson regression  model, or a Negative Binomial regression model, for example.

However, such data occur frequently in practice. So, what options are open to us?

EconAcademics.org

This, last evening, from Christian Zimmermann, at the Federal Reserve Bank of St. Louis:

"Dear Blogger,

Congratulations, you made the list!

The Federal Reserve Bank of St. Louis is launching a blog aggregator, EconAcademics.org, to highlight and promote the discussion of  economics research. Your blog is part of this effort. This email explains why and how you can help promote the discussion of economic research in the blogosphere.

EconAcademics.org lives at http://econacademics.org and aggregates blog posts that discuss economic research. The aggregator looks through blog posts for a link to some research indexed on a RePEc service, currently EconPapers, IDEAS and NEP. IDEAS then also links back from the abstract page to the blog posts."

Another nice service from the providers of RePEc!

© 2012, David E. Giles