Juea Pearl's Causal Analysis Blog

Judea Pearl is a big name in causal analysis. His book Causality: Models, Reasoning and Inference is already a classic (a second edition is scheduled to come out in October 2009). In contrast to Donald Rubin, Judea Pearl takes a graphical approach to causality. The basic idea is to represent conditional dependencies in a path diagram. Professor Pearl writes a blog on causality that can be found here.

U.S. Foreclosures Rise 15% in First Half of 2009

The following is a brief summary of a recent article in the The Dallas Morning News. It speaks for itself.

The number of U.S. households on the verge of losing their homes soared by nearly 15 percent in the first half of the year. The foreclosure crisis affected more than 1.5 million homes. More than 336,000 households received at least one foreclosure-related notice in June. It was the fourth-straight month in which more than 300,000 households receiving a foreclosure filing. Experts don't expect foreclosures to peak until the middle of next year.

Markov chain Monte Carlo Methods Are Important in many Areas of Applied Econometrics

Using Bayesian methods, it is often important to be able to sample from a posterior distribution. Markov chain Monte Carlo (MCMC) methods are of particular importance. Popular random walk MCMC methods include the Metropolis-Hastings algorithm and Gibbs sampling (which is a special case of Metropolis-Hastings).

Why would we need MCMC methods? The most common application is the numerical calculation of mutli-dimensional integrals. Numerical integration is key in many areas of applied econometrics. A simple example is the random effects binary model. The dependent variable takes only values 0 and 1 and the individual effects are normally distributed. This model is estimated with maximum likelihood estimation. The log-likelihood function consists of an integral for which no analytical solution exists. Thus, numerical methods are needed. In this case, integration is one-dimensional, so methods other than MCMC can be used. But it illustrates that even in basic econometric applications (that are not necessarily based on Bayesian methods) it is possible that numerical integration plays a key role.

Hierarchical Models and Andrew Gelman's Blog

Andrew Gelman is professor of statistics and political science at Columbia University. His book Data Analysis Using Regression and Multilevel/Hierarchical Models (with Jennifer Hill) is also quite important for empirical corporate finance. Hierarchical models are also known under the names mixed models, mixed effects models, multilevel models, random coefficients models, and variance components models.

Consider for simplicity linear models. Mixed linear models are related to what econometricians call the random effects model. The key difference is that random effects models specify only the intercept coefficient to be random. Richer models such as mixed linear models also permit the slope parameters to be random.

Professor Gelman also has a great blog where discusses many important concepts. See for example his post on how to think about instrumental variables when you get confused (and who doesn't sometimes get confused when trying to come up with good instruments). Professor Jackman has a related blog, the most interesting parts from FinanceStrudel's point of view are probably those on statistics and computing.

US Federal Deficit to Hit $ 2 Trillion

For the first time ever, the federal deficit of the US has topped $ 1 trillion. This number is breathtaking. But it even gets better: the deficit could grow to nearly $ 2 trillion this fall!

Everyone around the world holding US bonds is getting very nervous. Everybody is talking about China, but there are many other countries (e.g. Japan and several countries in the Middle East) sitting in the same boat. Just think about the Chiasso smuggling case. There is still hot debate whether or not the bonds and bills were genuine or fake. Meanwhile the unemployment rate in the US has climbed to 9.5%. Everybody who believes that the crisis will be over soon should think about these numbers.

RGE Monitor

RGE Monitor focuses on macroeconomic, financial, geo-strategic, and geo-political news. It was founded in 2004 by Nouriel Roubini (a.k.a. Dr. Doom), a professor of economics at NYU.

Al Roth's Market Design Blog

There is a very interesting blog on market design by Al Roth. He discusses current news from a mechanism design perspective.

Posted by Matthias.