A Bayesian Consumption Function

What the title of this post is supposed to mean is: "Estimating a simple aggregate consumption function using Bayesian regression analysis".

In a recent post I mentioned my long-standing interest in Bayesian Econometrics. When I teach this material I usually include a simple application that involves estimating a consumption function using U.S. time-series data. I used to have this coded up for the SHAZAM package. EViews isn't appropriate as it doesn't include a numerical integration routine.

You could use BUGS, or some other package, but it's nice to see what is going on, step-by-step, when you're encountering this stuff for the first time.

The other day, I thought, "It's time to code this up in R". So, here we are!

LeaRning R

If, like me, you're still an R novice, you'll no doubt find this post on Pairach Piboonrungroj's blog extremely helpful. Among other things, Pairach provides links to 20 40 "R tutorials". It's a really nice resource!

H.T. to David Smith for posting about this on the Revolutions blog.

© 2012, David E. Giles

Grad. Students' Projects

It's that time of year again - classes are over and I'm grading exams and projects. This term I was teaching an introductory Economic Statistics course and an elective grad. Econometrics course. I've graded the latter exams., but I'm buried in about 200 scripts for the undergraduate course. I've also got the term projects for the graduate course to go through. They look really good!

Natural Resources and Canadian Real Income Growth

In today's issue of The Daily, Statistics Canada has released a terrific study titled "The Role of Natural Resources in Real Income Growth". The paper's authors are John Baldwin and (former UVic. grad. student) Ryan McDonald. (H.T. to Ryan for bringing this to my attention.)

The study itself is extremely comprehensive, and provides some interesting new results about long-term economic growth in Canada. You can download the pdf version from here.

Drop-Down Menus for R

A few days ago, Andrew Barr had a great post on his blog. It was titled, "R is not just for nerds....it has drop-down menus!" You can bet that this one caught my eye when it was re-posted on R-Bloggers.

Briefly, Andrew takes us through the installation and basic use of the Java Gui for R (JGR) in conjunction with the Deducer package. Andrew noted in a previous post last month that a big advantage that JGR (say "Jaguar") has over the alternative GUI interfaces for R that are around is that it genuinely cross-platform, and will work in the same way on different operating systems.

I'm definitely going to be playing with it!

© 2012, David E. Giles

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